[{"id":1795022,"web_url":"http://patchwork.ozlabs.org/comment/1795022/","msgid":"","list_archive_url":null,"date":"2017-10-27T16:29:30","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":65878,"url":"http://patchwork.ozlabs.org/api/people/65878/","name":"Zack Weinberg","email":"zackw@panix.com"},"content":"On Tue, Sep 19, 2017 at 6:42 AM, Rafal Luzynski\n wrote:\n> Some languages (Slavic, Baltic, etc.) require a genitive case of the\n> month name when formatting a full date (with the day number) while\n> they require a nominative case when referring to the month standalone.\n> This requirement cannot be fulfilled without providing two forms for\n> each month name. From now it is precised that nl_langinfo(MON_1)\n> series (up to MON_12) and strftime(\"%B\") generate the month names in\n> the grammatical form used when the month forms part of a complete date.\n> If the grammatical form used when the month is named by itself is needed,\n> the new values nl_langinfo(ALTMON_1) (up to ALTMON_12) and\n> strftime(\"%OB\") are supported.\n\nI am reviewing this patch on the assumption that we have consensus for\nthe addition of this feature, with the semantics described above, and\nwithout any provision for old binaries (that is, the strings produced\nby %B will unconditionally change when locales that need two forms are\nupdated). Last call for objections.\n\nThe code changes look good to me, except for a few small concerns\nlisted below. Also, before landing I would like Joseph Myers (cc:ed)\nto positively confirm that this patch introduces no new ISO C or POSIX\nconformance issues, and will not tie our hands if this or a similar\nfeature is standardized in the future.\n\nI do not see any new tests anywhere in this patch series, and I found\na bug in the support for %OB in strptime (see below). To test this\nfeature we need at least one locale that actually uses it, so rather\nthan gate *these* patches on the addition of tests, I am asking you to\nwrite comprehensive tests and include them with the first patch you\nsubmit that adds altmon names to a locale. (This locale will have to\nbe one of the locales currently listed in localedata/Makefile as\navailable to tests (the test-input variable, I think); if none of them\nneed this feature, you may add one.)\n\n> [BZ #10871]\n> * locale/C-time.c: Add alternative month names, define them as the\n> same as mon explicitly.\n> * locale/categories.def: alt_mon and wide-alt_mon added.\n> * locale/langinfo.h: ALTMON_1 .. ALTMON_12 and similar contants\n> defined.\n> * locale/programs/ld-time.c: Alternative month names support\n> added, they are a copy of mon if not specified explicitly.\n> * locale/programs/locfile-kw.gperf: alt_mon defined.\n> * locale/programs/locfile-token.h: tok_alt_mon defined.\n> * localedata/tst-langinfo.c: Add tests for the new constants\n> ALTMON_1 .. ALTMON_12.\n> * time/strftime_l.c: %OB format for alternative month names\n> added.\n> * time/strptime_l.c: Alternative month names also recognized.\n\nA note on writing ChangeLog entries: we really do want you to list\nevery single new symbol that's part of the public API, not an\nabbreviated list -- that is:\n\n * locale/langinfo.h [__USE_GNU]: New public symbols ALTMON_1,\n ALTMON_2, ALTMON_3, ALTMON_4, ALTMON_5, ALTMON_6, ALTMON_7,\n ALTMON_8, ALTMON_9, ALTMON_10, ALTMON_11, ALTMON_12,\n _NL_WALTMON_1, _NL_WALTMON_2, _NL_WALTMON_3, _NL_WALTMON_4,\n _NL_WALTMON_5, _NL_WALTMON_6, _NL_WALTMON_7, _NL_WALTMON_8,\n _NL_WALTMON_9, _NL_WALTMON_10, _NL_WALTMON_11, _NL_WALTMON_12.\n\nWe want it spelled out like this because the ChangeLog exists to be\ngrepped. The idea is that someone five years from now who is having a\nweird problem involving _NL_WALTMON_9 should be able to search for\n_NL_WALTMON_9 and find the commit where it was introduced. If you\nabbreviate the list, that won't work.\n\nThe official instructions for writing ChangeLogs\n would\nhave you mention *every* new, changed, or deleted symbol for *every*\nfile where it appears, but I find that that bulks up the log entries\nwithout adding much benefit. It is enough, I think, to mention every\nfile, and (independently) every new, changed, or deleted symbol at\nleast once. That's enough for the hypothetical future person to find\nthe commit, and then they're going to look at the diffs.\n\n> diff --git a/ChangeLog b/ChangeLog\n> index 3b8e6c5..b0636ff 100644\n> --- a/ChangeLog\n> +++ b/ChangeLog\n\nAnother minor procedural thing: don't include diffs to the file\nChangeLog in commits sent for review, because if a reviewer wants to\napply the patch and tinker with it, they don't want to have to clean\nup the inevitable merge botch in the ChangeLog.\n\n> @@ -402,6 +404,20 @@ __strptime_internal (const char *rp, const char *fmt,\n> struct tm *tmp,\n> if (s.decided !=raw)\n> {\n> trp = rp;\n> +#ifdef _LIBC\n> + /* First check the alt month. */\n\nHere you are checking the alt-month name before the month name ...\n\n> @@ -428,6 +444,10 @@ __strptime_internal (const char *rp, const char *fmt,\n> struct tm *tmp,\n> if (s.decided != loc\n> && (((trp = rp, match_string (month_name[cnt], trp))\n> && trp > rp_longest)\n> +#ifdef _LIBC\n> + || ((trp = rp, match_string (alt_month_name[cnt], trp))\n> + && trp > rp_longest)\n> +#endif\n> || ((trp = rp, match_string (ab_month_name[cnt], trp))\n> && trp > rp_longest)))\n\n... and here you are checking the alt-month name after the month name.\nPlease make this consistent one way or the other. (I *think* the\norder of checks would only ever matter if the alt-month name for month\nX were the same as the month name for month Y, which would be a nasty\nambiguity in the relevant natural language -- but nasty ambiguities do\nactually happen in natural languages, so we need to be careful.)\n\nI wonder whether we actually need the #ifdef _LIBC here -- but that's\nan independent question (is this code _really_ still shared with\ngnulib, or have they diverged to the point where we should stop trying\nto keep the files in sync?) and you are not on the hook to answer it.\n\n> @@ -1015,6 +1035,10 @@ __strptime_internal (const char *rp, const char *fmt,\n> struct tm *tmp,\n> case 'O':\n> switch (*fmt++)\n> {\n> + case 'B':\n> + /* Undo the increment and continue. */\n> + fmt--;\n> + break;\n\nThis is subtly wrong, and I had to read a big chunk of the entire\nfunction to understand what it was meant to do and how it doesn't\nactually work. Please change to\n\n] case 'O':\n] switch (*fmt++)\n] {\n] + case 'B':\n] + /* Match month name. Reprocess as plain 'B'. */\n] + fmt--;\n] + goto start_over;\n\nYou will also need to remove the `#ifndef _NL_CURRENT` wrapping the\ndefinition of the `start_over` label, and change its comment:\n\n] -#ifndef _NL_CURRENT\n] - /* We need this for handling the `E' modifier. */\n] + /* In some cases, modifiers are handled by adjusting state and\n] + then restarting the switch statement below. */\n] start_over:\n] -#endif\n]\n] /* Make back up of current processing pointer. */\n\n(Am I seriously telling you to use `goto`? Yes, I am. This is the\nkind of abnormal control flow situation where a judicious `goto` is\n_easier_ to read than the alternative - what you had was meant to do\nthe same thing, but it did it by cycling the `while (*fmt != '\\0')`\nloop, which meant that all of the processing _above_ the outer switch\nstatement was repeated, which was both wrong (note the unconditional\n`++fmt` above \"We discard strftime modifiers\") and more fragile and\ncomplicated to understand than \"goto start_over\", which requires the\nreader only to search for the \"start_over\" label.)\n\nzw","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86484-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"L0hkWzl+\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yNq8k0Jt0z9t5c\n\tfor ;\n\tSat, 28 Oct 2017 03:29:41 +1100 (AEDT)","(qmail 32577 invoked by alias); 27 Oct 2017 16:29:36 -0000","(qmail 32190 invoked by uid 89); 27 Oct 2017 16:29:36 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; q=dns; s=default; b=N4ac\n\twhoyvMCLXRgKleVUhSpWZ1lELBHLlNMEWE7YlmZUP6O6pRlJ5KOkUhashFhq5gl/\n\t8i0TJbl8yb6NmSA5KzItykso5EgA48w4wDe2HgGfjH9w3LJky24zWbWwJwBdsnVY\n\t3dZgehcKgaOwuW6QjMSxHaOUJDZFuedrkp4Qvc0=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; s=default; bh=LhNNICKpqH\n\tCfULa2ax/QFa3woTs=; b=L0hkWzl+rOcpx/TXe+Y6BYIh5wk6w4wPk52ISoxHK5\n\t2VucCPIdsuTXbnOCb07nKrCdTeKE+d+LKM2fQGQrlFlJgsfb9EbM4ZnEeLl/afTj\n\tIVObCmoaWEHS09t+OmemHspTBHOcklcRetjuIQVtIe29JVC//71N/F/s6dOgEHZb\n\t4=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-23.9 required=5.0 tests=AWL, BAYES_00,\n\tGIT_PATCH_0, GIT_PATCH_1, GIT_PATCH_2, GIT_PATCH_3,\n\tRCVD_IN_SORBS_SPAM, RP_MATCHES_RCVD,\n\tSPF_PASS autolearn=ham version=3.3.2 spammy=cycling,\n\tabbreviated, seriously, month","X-HELO":"mailbackend.panix.com","X-Gm-Message-State":"AMCzsaW1jOyanYNCYSbQHIZ5y5cYnscbTGzJjDCHvx3EWYb0spnWkQ8J\n\tksf16YRO3YeaqsqjogrXDYxpk74ue0z9WMR297I=","X-Google-Smtp-Source":"ABhQp+Q53OxbDRHiL6hFNLw17Fz+CLScXdri7/r6Jbsq2VW3p/Pfwy48sBtRk4YFhcJzEkoy6VFAmOvw3Jf3tgDCzhk=","X-Received":"by 10.202.178.133 with SMTP id b127mr598769oif.261.1509121771296;\n\tFri, 27 Oct 2017 09:29:31 -0700 (PDT)","MIME-Version":"1.0","In-Reply-To":"<742475879.1094767.1505817734249@poczta.nazwa.pl>","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>","From":"Zack Weinberg ","Date":"Fri, 27 Oct 2017 12:29:30 -0400","X-Gmail-Original-Message-ID":"","Message-ID":"","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","To":"Rafal Luzynski ","Cc":"GNU C Library ,\n\tJoseph Myers ","Content-Type":"text/plain; charset=\"UTF-8\""}},{"id":1795038,"web_url":"http://patchwork.ozlabs.org/comment/1795038/","msgid":"","list_archive_url":null,"date":"2017-10-27T17:03:49","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":4349,"url":"http://patchwork.ozlabs.org/api/people/4349/","name":"Joseph Myers","email":"joseph@codesourcery.com"},"content":"On Fri, 27 Oct 2017, Zack Weinberg wrote:\n\n> The code changes look good to me, except for a few small concerns\n> listed below. Also, before landing I would like Joseph Myers (cc:ed)\n> to positively confirm that this patch introduces no new ISO C or POSIX\n> conformance issues, and will not tie our hands if this or a similar\n> feature is standardized in the future.\n\nI believe this patch properly handles namespace issues and avoids any \nproblems with improperly changing enum values or making them improperly \ndepend on feature test macros defined. I believe it properly deals with \nallowing existing locales using existing POSIX features to continue to \nwork by copying the month values as needed.\n\nThere should be a GCC bug filed, if there isn't one already, for allowing \n%OB / %Ob in strftime formats (with appropriate warnings in ISO C pedantic \nmodes about being an extension). Tests may need to disable errors for \nformat warnings when built with a GCC version without that support.\n\n> > @@ -1015,6 +1035,10 @@ __strptime_internal (const char *rp, const char *fmt,\n> > struct tm *tmp,\n> > case 'O':\n> > switch (*fmt++)\n> > {\n> > + case 'B':\n> > + /* Undo the increment and continue. */\n> > + fmt--;\n> > + break;\n> \n> This is subtly wrong, and I had to read a big chunk of the entire\n> function to understand what it was meant to do and how it doesn't\n> actually work. Please change to\n> \n> ] case 'O':\n> ] switch (*fmt++)\n> ] {\n> ] + case 'B':\n> ] + /* Match month name. Reprocess as plain 'B'. */\n> ] + fmt--;\n> ] + goto start_over;\n\nAlso, the POSIX proposal accepted for issue 8 has strptime %Ob and \nstrptime %OB handled the same, despite that POSIX proposal not including \nstrftime %Ob. (This is only an argument for %Ob handling for strptime to \ngo in this patch rather than the later one adding alternative abbreviated \nmonths, however.)","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86487-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"Hr1jTQv5\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yNqwR4hxrz9t4B\n\tfor ;\n\tSat, 28 Oct 2017 04:04:07 +1100 (AEDT)","(qmail 32738 invoked by alias); 27 Oct 2017 17:04:02 -0000","(qmail 32060 invoked by uid 89); 27 Oct 2017 17:04:01 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; q=dns; s=default; b=J+sUT\n\taCg4k28PG0OjhXfisr3wJQoH9XBcNGfuj0kVUfW9YbMli6BCBE5grQ/L9nDHXsFN\n\tfQBlM8xtc5iFsp59MihE+UnMGczAwEq/5NSk2l9uR4nzAM9Phoo8A0jt4awLlmHf\n\tvn8W/R78sb+Ca0QNMBB2ntQgrwPjMX2KYQpxCU=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; s=default; bh=yDfmEwe8Li7\n\tnOyuFSB4rGPSe4TU=; b=Hr1jTQv5o3clvUpuL183jpCbiyegBOH8lXvlJgWyUnF\n\tgPjXy3BpYo7PgdVSQlvpR3vDTEia9ycj+oIfmCJKGONCTjipafakO3xT5rIzESuo\n\t8igLXdkCzMjT7yKK+kwj4GGTbdL9MyyfCjyjQ0W9h3Zz2fGEslfwJwOQxb/DmBzc\n\t=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-2.0 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE, SPF_PASS,\n\tURIBL_RED autolearn=ham version=3.3.2 spammy=month, our","X-HELO":"relay1.mentorg.com","Date":"Fri, 27 Oct 2017 17:03:49 +0000","From":"Joseph Myers ","To":"Zack Weinberg ","CC":"Rafal Luzynski , GNU C Library\n\t","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","In-Reply-To":"","Message-ID":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t","User-Agent":"Alpine 2.20 (DEB 67 2015-01-07)","MIME-Version":"1.0","Content-Type":"text/plain; charset=\"US-ASCII\"","X-ClientProxiedBy":"svr-ies-mbx-01.mgc.mentorg.com (139.181.222.1) To\n\tSVR-IES-MBX-03.mgc.mentorg.com (139.181.222.3)"}},{"id":1795040,"web_url":"http://patchwork.ozlabs.org/comment/1795040/","msgid":"","list_archive_url":null,"date":"2017-10-27T17:09:56","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":65878,"url":"http://patchwork.ozlabs.org/api/people/65878/","name":"Zack Weinberg","email":"zackw@panix.com"},"content":"On Fri, Oct 27, 2017 at 1:03 PM, Joseph Myers wrote:\n> On Fri, 27 Oct 2017, Zack Weinberg wrote:\n>\n>> The code changes look good to me, except for a few small concerns\n>> listed below. Also, before landing I would like Joseph Myers (cc:ed)\n>> to positively confirm that this patch introduces no new ISO C or POSIX\n>> conformance issues, and will not tie our hands if this or a similar\n>> feature is standardized in the future.\n>\n> I believe this patch properly handles namespace issues and avoids any\n> problems with improperly changing enum values or making them improperly\n> depend on feature test macros defined. I believe it properly deals with\n> allowing existing locales using existing POSIX features to continue to\n> work by copying the month values as needed.\n\nThanks.\n\n> There should be a GCC bug filed, if there isn't one already, for allowing\n> %OB / %Ob in strftime formats (with appropriate warnings in ISO C pedantic\n> modes about being an extension). Tests may need to disable errors for\n> format warnings when built with a GCC version without that support.\n\nI don't see any such bug in GCC bugzilla. Would you mind filing one?\n\n> Also, the POSIX proposal accepted for issue 8 has strptime %Ob and\n> strptime %OB handled the same, despite that POSIX proposal not including\n> strftime %Ob. (This is only an argument for %Ob handling for strptime to\n> go in this patch rather than the later one adding alternative abbreviated\n> months, however.)\n\nSince the entire patch series should be pushed at once anyway, let's\nnot make additional work for Rafal.\n\nCould you point me at the accepted POSIX proposal, please?\n\nzw","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86488-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"IV/a4Q3i\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yNr3L6QY5z9s0Z\n\tfor ;\n\tSat, 28 Oct 2017 04:10:06 +1100 (AEDT)","(qmail 29610 invoked by alias); 27 Oct 2017 17:10:01 -0000","(qmail 29597 invoked by uid 89); 27 Oct 2017 17:10:00 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; q=dns; s=default; b=lUVM\n\tXbAXM81volIp9ExrxJyVah9lbYjEoGS/5gO3UeISTcTH5HFgYytYd6wEPJK5Z1aW\n\tr4nUkpXZdUL9iuiMHdbWibMgFqqUA+OK3mxzVHRRDSRyle7y2bTLGg0HxIvyQu5E\n\tnzcFdZe6bf3un/Nk+pe5zNX7LvOFBXAOEWCoFDs=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; s=default; bh=HESjU7fvDn\n\tt1STPb6J1mqqAPYsw=; b=IV/a4Q3idgUBPrTkMP9i0OzoQXCZVTQwThXtTPSt+y\n\t3PQv78GB24im2Pljtq/Aj2q4U7qYZXU1bP8+Hxbe0N458MAZzAnfooQZ1CHfO04n\n\tj5Y9KhG6kx/HiUUbpU1WB02WR+5gSnos5jpzccy/BKMArmyEOuEEERqidxRdi2+m\n\tA=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.5 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_SORBS_SPAM, RP_MATCHES_RCVD, SPF_PASS,\n\tURIBL_RED autolearn=no version=3.3.2 spammy=month, our","X-HELO":"mailbackend.panix.com","X-Gm-Message-State":"AMCzsaXi+68pY+f55B8b/YeSiKzBKOT/OqQoRjYxo1a7xRZrTy+YjKgx\n\tGw+0iA0aAqXLKZkMPKTFAga05GY3scnk32Sxvbo=","X-Google-Smtp-Source":"ABhQp+RmQDszprxs33ahRShzFO+UB17pilEn4odfvhWXwXIoVdOBbVbz6lhvprSN/NU1BFYcEsBFrAgofRjEmbgFVdI=","X-Received":"by 10.202.56.196 with SMTP id f187mr556717oia.285.1509124197411; \n\tFri, 27 Oct 2017 10:09:57 -0700 (PDT)","MIME-Version":"1.0","In-Reply-To":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t","From":"Zack Weinberg ","Date":"Fri, 27 Oct 2017 13:09:56 -0400","X-Gmail-Original-Message-ID":"","Message-ID":"","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","To":"Joseph Myers ","Cc":"Rafal Luzynski , \n\tGNU C Library ","Content-Type":"text/plain; charset=\"UTF-8\""}},{"id":1795044,"web_url":"http://patchwork.ozlabs.org/comment/1795044/","msgid":"","list_archive_url":null,"date":"2017-10-27T17:19:27","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":4349,"url":"http://patchwork.ozlabs.org/api/people/4349/","name":"Joseph Myers","email":"joseph@codesourcery.com"},"content":"On Fri, 27 Oct 2017, Zack Weinberg wrote:\n\n> > There should be a GCC bug filed, if there isn't one already, for allowing\n> > %OB / %Ob in strftime formats (with appropriate warnings in ISO C pedantic\n> > modes about being an extension). Tests may need to disable errors for\n> > format warnings when built with a GCC version without that support.\n> \n> I don't see any such bug in GCC bugzilla. Would you mind filing one?\n\nhttps://gcc.gnu.org/bugzilla/show_bug.cgi?id=82752 filed.\n\n> > Also, the POSIX proposal accepted for issue 8 has strptime %Ob and\n> > strptime %OB handled the same, despite that POSIX proposal not including\n> > strftime %Ob. (This is only an argument for %Ob handling for strptime to\n> > go in this patch rather than the later one adding alternative abbreviated\n> > months, however.)\n> \n> Since the entire patch series should be pushed at once anyway, let's\n> not make additional work for Rafal.\n> \n> Could you point me at the accepted POSIX proposal, please?\n\nhttp://austingroupbugs.net/view.php?id=258 (issue8 tag = for next major \nrevision).","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86489-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"p+6uH4a0\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yNrGR6Tpxz9rvt\n\tfor ;\n\tSat, 28 Oct 2017 04:19:43 +1100 (AEDT)","(qmail 44656 invoked by alias); 27 Oct 2017 17:19:37 -0000","(qmail 44647 invoked by uid 89); 27 Oct 2017 17:19:37 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; q=dns; s=default; b=bzxki\n\twjTIhlpfoNtL2WTVxjDYiUg/Aj1cbB0OXmPMpTK4fchapfj3+8Kc3SrjUzTEb3CY\n\tuCdnDre1hvgw6ubKXwu7LFREvYNAlxIYALjF757iW4mr3xf6Lo9QggqNpZ+stODJ\n\tSFRY3gtrTIe2ZYjf6YFQqBsaLq5aZd2S0VC/pg=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; s=default; bh=K7eJIAWXXlB\n\tbXiVwSkMvr/E82gk=; b=p+6uH4a08KOnhsFiRsJ/wAO8Wfo78lAzTr+cTDfg0HY\n\tfHqR7Nr1yhKUIFKn9R08D9JqiBMpeTpkJhWQjTl5ceE9gOYvOum6NCwbbNBeVTJH\n\tY8n88UR9g82siXJh/2XCGShiElPqKAJVcPTa5FhufooKqMOVfwKNBVzDvTe92d1U\n\t=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-2.0 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE, SPF_PASS,\n\tURIBL_RED autolearn=ham version=3.3.2 spammy=month, UD:php","X-HELO":"relay1.mentorg.com","Date":"Fri, 27 Oct 2017 17:19:27 +0000","From":"Joseph Myers ","To":"Zack Weinberg ","CC":"Rafal Luzynski , GNU C Library\n\t","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","In-Reply-To":"","Message-ID":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t\n\t","User-Agent":"Alpine 2.20 (DEB 67 2015-01-07)","MIME-Version":"1.0","Content-Type":"text/plain; charset=\"US-ASCII\"","X-ClientProxiedBy":"svr-ies-mbx-01.mgc.mentorg.com (139.181.222.1) To\n\tSVR-IES-MBX-03.mgc.mentorg.com (139.181.222.3)"}},{"id":1795055,"web_url":"http://patchwork.ozlabs.org/comment/1795055/","msgid":"","list_archive_url":null,"date":"2017-10-27T17:51:26","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":65878,"url":"http://patchwork.ozlabs.org/api/people/65878/","name":"Zack Weinberg","email":"zackw@panix.com"},"content":"On Fri, Oct 27, 2017 at 1:19 PM, Joseph Myers wrote:\n>\n> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82752 filed.\n\nThanks. I added a note about strptime.\n\n> http://austingroupbugs.net/view.php?id=258 (issue8 tag = for next major\n> revision).\n\nHmm. Should we, somehow, tell them about the need for %Ob?\n\nzw","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86491-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"BZduw4ZM\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yNrzL3jwJz9t3H\n\tfor ;\n\tSat, 28 Oct 2017 04:51:42 +1100 (AEDT)","(qmail 127043 invoked by alias); 27 Oct 2017 17:51:31 -0000","(qmail 126999 invoked by uid 89); 27 Oct 2017 17:51:31 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; q=dns; s=default; b=dFfV\n\tU5ItzqxiGmp82Z2egkRkqfjqTsppLk3eVjg6bE7en8xz5IA+4LmkQLCmTgxxE2sW\n\tF/yT7JbNb7qASc2rT3bYYqX6IB8paDTPyZhq0puE1hB8S4yI4HOaVIXUu5WDa1gW\n\tKOJDyVEk4oRmhkbAAj8RBq8JarOKboOFOXPBmIM=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; s=default; bh=WtYuRSAyQa\n\tdl0BKXsnvy1SkfDpA=; b=BZduw4ZMI3IU5uMw1kqUu66BIWtkuhDaDiosT1ytRL\n\tUA8c8eldNHDgijT75dTbobilnSqKfNBqjaFcuB6n727byv7DBSHk76Bt80z0jYO0\n\tT1yg6dknB4EK5l3+BsYR6W1bosrMxhrHH2h8x2/JzwhuEH845Rw5WrYopvmySXIM\n\tU=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.6 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_SORBS_SPAM, RP_MATCHES_RCVD, SPF_PASS,\n\tURIBL_RED autolearn=no version=3.3.2 spammy=month, UD:php","X-HELO":"mailbackend.panix.com","X-Gm-Message-State":"AMCzsaVIH/Ufm2bKGtr5QJXUTBU7gGbinCmyBBh8QjGIkmWznOSk2G7T\n\tX8rsK68YU/NoCDvbQLw5WZmwCuUwIfsR/ectaho=","X-Google-Smtp-Source":"ABhQp+SmuBYUdL0PqEpCvYiibaPE1Pt+8XTIqa9zxyundMbLtVnHo9vAcXe91VmKcm6awZlpuAvPMSHS8DZdXJchj30=","X-Received":"by 10.157.67.42 with SMTP id s39mr817252ote.243.1509126687355;\n\tFri, 27 Oct 2017 10:51:27 -0700 (PDT)","MIME-Version":"1.0","In-Reply-To":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t\n\t\n\t","From":"Zack Weinberg ","Date":"Fri, 27 Oct 2017 13:51:26 -0400","X-Gmail-Original-Message-ID":"","Message-ID":"","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","To":"Joseph Myers ","Cc":"Rafal Luzynski , \n\tGNU C Library ","Content-Type":"text/plain; charset=\"UTF-8\""}},{"id":1795089,"web_url":"http://patchwork.ozlabs.org/comment/1795089/","msgid":"","list_archive_url":null,"date":"2017-10-27T19:25:01","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":4349,"url":"http://patchwork.ozlabs.org/api/people/4349/","name":"Joseph Myers","email":"joseph@codesourcery.com"},"content":"On Fri, 27 Oct 2017, Zack Weinberg wrote:\n\n> On Fri, Oct 27, 2017 at 1:19 PM, Joseph Myers wrote:\n> >\n> > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82752 filed.\n> \n> Thanks. I added a note about strptime.\n\nGCC doesn't have strptime format checking.\n\n> > http://austingroupbugs.net/view.php?id=258 (issue8 tag = for next major\n> > revision).\n> \n> Hmm. Should we, somehow, tell them about the need for %Ob?\n\nYes, probably. I believe anyone can create an account to file or comment \non bugs.","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86494-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"r5vUgayv\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yNv3R3CDSz9t4X\n\tfor ;\n\tSat, 28 Oct 2017 06:25:23 +1100 (AEDT)","(qmail 20284 invoked by alias); 27 Oct 2017 19:25:16 -0000","(qmail 20243 invoked by uid 89); 27 Oct 2017 19:25:12 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; q=dns; s=default; b=Qh8Gz\n\tD1uCgvnDD2N5Cr4xk8Rm/qC0KWhuISJaLvMB4NpQTLA4gs9z8RpMk6j4ek/mOv0f\n\tINxvpr9YaXCyPT7WvPKjUU7yJyq32CueCgYbSRJ90s6wV0jq201hsMYjUCkPR6nd\n\toVAWR7ox1J9cvnRh8OnrYcJrvYBElwEgZxE6Oo=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; s=default; bh=puHKGxVuBiI\n\t8GHwmj4WX1XW7vE8=; b=r5vUgayv3Gmm2H8HhDCVbeWSU7m4mznDpI6LzOTvm3x\n\thBjNAeNBdrL7Ix4KnxNf4umcpG5hugP2kt6dFQ2thzgYHW7ifHdZQLKijOYiaCkD\n\tR19V2WElEM8Tj0uq1SLKwq8PKyf72Uhv6bkLGiYtnxIGOsYYmOqinDaOfgA5bNBk\n\t=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-2.0 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE, SPF_PASS,\n\tURIBL_RED autolearn=ham version=3.3.2 spammy=month, UD:php","X-HELO":"relay1.mentorg.com","Date":"Fri, 27 Oct 2017 19:25:01 +0000","From":"Joseph Myers ","To":"Zack Weinberg ","CC":"Rafal Luzynski , GNU C Library\n\t","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","In-Reply-To":"","Message-ID":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t\n\t\n\t\n\t","User-Agent":"Alpine 2.20 (DEB 67 2015-01-07)","MIME-Version":"1.0","Content-Type":"text/plain; charset=\"US-ASCII\"","X-ClientProxiedBy":"svr-ies-mbx-01.mgc.mentorg.com (139.181.222.1) To\n\tSVR-IES-MBX-03.mgc.mentorg.com (139.181.222.3)"}},{"id":1795106,"web_url":"http://patchwork.ozlabs.org/comment/1795106/","msgid":"","list_archive_url":null,"date":"2017-10-27T20:16:02","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":65878,"url":"http://patchwork.ozlabs.org/api/people/65878/","name":"Zack Weinberg","email":"zackw@panix.com"},"content":"On Fri, Oct 27, 2017 at 3:25 PM, Joseph Myers wrote:\n> On Fri, 27 Oct 2017, Zack Weinberg wrote:\n>>> On Fri, Oct 27, 2017 at 1:19 PM, Joseph Myers wrote:\n>> > http://austingroupbugs.net/view.php?id=258 (issue8 tag = for next major\n>> > revision).\n>>\n>> Hmm. Should we, somehow, tell them about the need for %Ob?\n>\n> Yes, probably. I believe anyone can create an account to file or comment\n> on bugs.\n\nI have filed http://austingroupbugs.net/view.php?id=1166 .\n\nzw","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86495-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"M8xGQk1m\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yNwB46r49z9t3f\n\tfor ;\n\tSat, 28 Oct 2017 07:16:12 +1100 (AEDT)","(qmail 119290 invoked by alias); 27 Oct 2017 20:16:07 -0000","(qmail 118968 invoked by uid 89); 27 Oct 2017 20:16:06 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; q=dns; s=default; b=h5GE\n\t+wBeyDVGIUkUEFQxyRHVRPLuZq2hvsTtHCKNU8forVJh5oQEvtZZfFGnCnyPkKu0\n\tmpRJ826WoS/L+L866nDGOYejlWpoSJ8XRs/QLxvztENWRORmlulG4Q0eUTNlwKnn\n\tsZBVItXUly99x/vm/61MLFaF5elu2fQzPBt3k58=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; s=default; bh=2zsAzdUhUW\n\tHY0WhVOmLrRCxGEgA=; b=M8xGQk1mINoWv8NTJbyO/GM5pZgCJx21HfBGs9AVA8\n\tZJyHJTPz9j9N9KYTygq6dkBcPtL66wP1l3nORBq98ekVrSWyEqHBDQWYFVzugXB4\n\tAaLCnmeDU8Jn2WmyL9O/q3dko591kKw7Opu2hnffV16uRr4rZ2vxtwPWNR1CeIFW\n\tA=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.6 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_SORBS_SPAM, RP_MATCHES_RCVD, SPF_PASS,\n\tURIBL_RED autolearn=no version=3.3.2 spammy=1166, month,\n\tUD:php","X-HELO":"mailbackend.panix.com","X-Gm-Message-State":"AMCzsaXC3O8jiBTyczWe9QWzscJPNczePVU8seBAQxSEl9radkq+wTZi\n\tquo0leR/q/0+SZD+rwYgcmeK1jL3+RHPTNW4Ll8=","X-Google-Smtp-Source":"ABhQp+RWF83SJpsHpnrckvLUwLlESmoUuoCbe+atGzQOOg5vag08Vp23YAIS/ZbAkQn+qAw1NxPI14LX+LptOttFrAg=","X-Received":"by 10.157.46.90 with SMTP id c26mr1052873otd.433.1509135362490; \n\tFri, 27 Oct 2017 13:16:02 -0700 (PDT)","MIME-Version":"1.0","In-Reply-To":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t\n\t\n\t\n\t\n\t","From":"Zack Weinberg ","Date":"Fri, 27 Oct 2017 16:16:02 -0400","X-Gmail-Original-Message-ID":"","Message-ID":"","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","To":"Joseph Myers ","Cc":"Rafal Luzynski , \n\tGNU C Library ","Content-Type":"text/plain; charset=\"UTF-8\""}},{"id":1795109,"web_url":"http://patchwork.ozlabs.org/comment/1795109/","msgid":"","list_archive_url":null,"date":"2017-10-27T20:30:07","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":4349,"url":"http://patchwork.ozlabs.org/api/people/4349/","name":"Joseph Myers","email":"joseph@codesourcery.com"},"content":"On Fri, 27 Oct 2017, Zack Weinberg wrote:\n\n> On Fri, Oct 27, 2017 at 3:25 PM, Joseph Myers wrote:\n> > On Fri, 27 Oct 2017, Zack Weinberg wrote:\n> >>> On Fri, Oct 27, 2017 at 1:19 PM, Joseph Myers wrote:\n> >> > http://austingroupbugs.net/view.php?id=258 (issue8 tag = for next major\n> >> > revision).\n> >>\n> >> Hmm. Should we, somehow, tell them about the need for %Ob?\n> >\n> > Yes, probably. I believe anyone can create an account to file or comment\n> > on bugs.\n> \n> I have filed http://austingroupbugs.net/view.php?id=1166 .\n\nI note that includes ABALTMON_* constants in langinfo.h, whereas the \npresent patch series has ALTMON_* constants but no ABALTMON_* names.","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86496-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"AXVwpNYu\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yNwVq5bJyz9t5R\n\tfor ;\n\tSat, 28 Oct 2017 07:30:43 +1100 (AEDT)","(qmail 109735 invoked by alias); 27 Oct 2017 20:30:37 -0000","(qmail 109620 invoked by uid 89); 27 Oct 2017 20:30:28 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; q=dns; s=default; b=Q52IT\n\tvF4NIpmPANFM8dzhp2YhOKbRLKJv3jG9GwbZ8U6iXYwF6agQQuJV6uSbAkDi4y+w\n\t/k1CaoCpOWP1zyYcBNpb5KheTNuwwkbx8iq5ddU1kPhq5d297orHy6QfzumOvMFx\n\tPguQma8mdGtL9i4Sn7n2RIv1yTLNQBw/A5j12M=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; s=default; bh=mD826VfzIQT\n\tONdvptpVkQBImDVw=; b=AXVwpNYuvDpiMzktr8dcELoCktJ4OHV0LRtIUKKNQOU\n\tx22FsDvKonOsxyC3pWESLdHP4MMOXxvGCW4IGvOhwtUFcIvo6fzMmyhw6iyL2ZHf\n\tB9wcyufddtip5R0k6TxawVpMceQRUs/CZp3t3yUxMsFwuT4PrURbYe6pPO2tw+V0\n\t=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-2.0 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE, SPF_PASS,\n\tURIBL_RED autolearn=ham version=3.3.2 spammy=1166, month,\n\tUD:php","X-HELO":"relay1.mentorg.com","Date":"Fri, 27 Oct 2017 20:30:07 +0000","From":"Joseph Myers ","To":"Zack Weinberg ","CC":"Rafal Luzynski , GNU C Library\n\t","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","In-Reply-To":"","Message-ID":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t\n\t\n\t\n\t\n\t\n\t","User-Agent":"Alpine 2.20 (DEB 67 2015-01-07)","MIME-Version":"1.0","Content-Type":"text/plain; charset=\"US-ASCII\"","X-ClientProxiedBy":"svr-ies-mbx-01.mgc.mentorg.com (139.181.222.1) To\n\tSVR-IES-MBX-03.mgc.mentorg.com (139.181.222.3)"}},{"id":1795111,"web_url":"http://patchwork.ozlabs.org/comment/1795111/","msgid":"","list_archive_url":null,"date":"2017-10-27T20:36:26","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":65878,"url":"http://patchwork.ozlabs.org/api/people/65878/","name":"Zack Weinberg","email":"zackw@panix.com"},"content":"On Fri, Oct 27, 2017 at 4:30 PM, Joseph Myers wrote:\n> On Fri, 27 Oct 2017, Zack Weinberg wrote:\n>\n>> On Fri, Oct 27, 2017 at 3:25 PM, Joseph Myers wrote:\n>> > On Fri, 27 Oct 2017, Zack Weinberg wrote:\n>> >>> On Fri, Oct 27, 2017 at 1:19 PM, Joseph Myers wrote:\n>> >> > http://austingroupbugs.net/view.php?id=258 (issue8 tag = for next major\n>> >> > revision).\n>> >>\n>> >> Hmm. Should we, somehow, tell them about the need for %Ob?\n>> >\n>> > Yes, probably. I believe anyone can create an account to file or comment\n>> > on bugs.\n>>\n>> I have filed http://austingroupbugs.net/view.php?id=1166 .\n>\n> I note that includes ABALTMON_* constants in langinfo.h, whereas the\n> present patch series has ALTMON_* constants but no ABALTMON_* names.\n\nRafal's patches have _NL_ABALTMON_*. I am not clear on when we\nactually want _NL_ prefixes on those constants, but I figured it was\nsimpler to write the proposal with no prefixes and the Austin Group\ncan add them if they want them.\n\nzw","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86497-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"YqD1cqhR\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yNwdc33l6z9t42\n\tfor ;\n\tSat, 28 Oct 2017 07:36:36 +1100 (AEDT)","(qmail 122477 invoked by alias); 27 Oct 2017 20:36:30 -0000","(qmail 122468 invoked by uid 89); 27 Oct 2017 20:36:30 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; q=dns; s=default; b=e3Zz\n\tx3LuNyeomtDqWfbHFotFBX5e6ZqQDYA9ITHBWAiExdPl4thfnSKUpzO9pqSMUPjb\n\thNV6HGQTIKJGhaj+tDPUyaRWdrlACtBBTHLa7Yml/dUUlAi6tHTHSotglj0Env6x\n\t+unnhf8DSdjY0hPKBANgDmoJ2igTlFUEo2GIKEU=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; s=default; bh=tqviyiyPWl\n\tQqRFVp50Etq5V17fI=; b=YqD1cqhRINwwJ0/6xyxSLsrWPTAtLWGwSkjWgV/3MO\n\t4QyrwdaEMGz9vufK7YEnmiute8LoK4RQUL2mBJ/PbEb6mPG6+ZrZpMdj4lUJs4ld\n\tfhG328u64jDLy0kqvzn8F1zUOwpqeF558WeOK8pt/7wtZQgzdhycz8rEGeYaKFuV\n\to=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.5 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_SORBS_SPAM, RP_MATCHES_RCVD, SPF_PASS,\n\tURIBL_RED autolearn=no version=3.3.2 spammy=Austin, austin,\n\tmonth, UD:php","X-HELO":"mailbackend.panix.com","X-Gm-Message-State":"AMCzsaX5Dki9dhjHoWCN/uL1R+OEqiMqTx86x4xMb9t2Rj2ugesjfNVN\n\tKxan73ZhF6TMWET5f7po/FFYpdtQiXPPOkTEViY=","X-Google-Smtp-Source":"ABhQp+QwB90YALI1tpu8FEDIV019wtbLcSzBLrO7GhY+AXc5oGN9aDc4vFinLWkAG/2JawaldrpwAOytkDWTVlLGI/I=","X-Received":"by 10.157.56.144 with SMTP id p16mr981073otc.224.1509136586593; \n\tFri, 27 Oct 2017 13:36:26 -0700 (PDT)","MIME-Version":"1.0","In-Reply-To":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t\n\t\n\t\n\t\n\t\n\t\n\t","From":"Zack Weinberg ","Date":"Fri, 27 Oct 2017 16:36:26 -0400","X-Gmail-Original-Message-ID":"","Message-ID":"","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","To":"Joseph Myers ","Cc":"Rafal Luzynski , \n\tGNU C Library ","Content-Type":"text/plain; charset=\"UTF-8\""}},{"id":1795842,"web_url":"http://patchwork.ozlabs.org/comment/1795842/","msgid":"<561657d6-c571-9f94-8ae9-277c888bb70f@redhat.com>","list_archive_url":null,"date":"2017-10-30T11:21:02","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":14312,"url":"http://patchwork.ozlabs.org/api/people/14312/","name":"Florian Weimer","email":"fweimer@redhat.com"},"content":"On 10/27/2017 06:29 PM, Zack Weinberg wrote:\n> I am reviewing this patch on the assumption that we have consensus for\n> the addition of this feature, with the semantics described above, and\n> without any provision for old binaries (that is, the strings produced\n> by %B will unconditionally change when locales that need two forms are\n> updated). Last call for objections.\n\nI still think the approach of redefining the meaning of %B is wrong and \nwill do more harm than good, but it's not a sustained objection.\n\n(This is not a matter of supporting old binaries, I think the compat \nsymbol approach would be even worse.)\n\nThanks,\nFlorian","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86518-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"OLz4G1O+\"; dkim-atps=neutral","sourceware.org; auth=none","ext-mx04.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx04.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=fweimer@redhat.com"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yQX9R1QTmz9t3Z\n\tfor ;\n\tMon, 30 Oct 2017 22:21:15 +1100 (AEDT)","(qmail 60758 invoked by alias); 30 Oct 2017 11:21:07 -0000","(qmail 60745 invoked by uid 89); 30 Oct 2017 11:21:06 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=VLX4Yq0nCpuKvKwc\n\tEihvkdZH/Z408s+hyBmFtoiesBZARvwZ7poKejZ1JLDetxFqDOUsHhe5/M7yc8nB\n\t/lD77AYKmm0DKrBT3OGC29BPzzRWWQt3VKyHvugUH9HjGTr+u8C+uq+7op7Rxxoz\n\tHrJQvLitTCOwdO6ljz0YO3QVps4=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=dMdaHHl+kBJ/qLRHylK8Kp\n\tL07eY=; b=OLz4G1O+oy6b5yZX6/o+s+afvi4X03gWWlxC6verbWd7+BXQE12iNq\n\tXSZGkPlaioCA8qJzEa9+pq4nYGl3r1Z6Z0r0bfxS56WVX8MUh1Ut/R/vawif5iWW\n\tG6VFVZHQ+YakedKkSTVVeQX69Gz8pp0+jBpPryhd3ll4ve+rB+wN4=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.9 required=5.0 tests=BAYES_00,\n\tRP_MATCHES_RCVD,\n\tSPF_HELO_PASS autolearn=ham version=3.3.2 spammy=HTo:D*panix.com,\n\tsustained, month","X-HELO":"mx1.redhat.com","DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com 75D607EA92","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","To":"Zack Weinberg ,\n\tRafal Luzynski ","Cc":"GNU C Library ,\n\tJoseph Myers ","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t","From":"Florian Weimer ","Message-ID":"<561657d6-c571-9f94-8ae9-277c888bb70f@redhat.com>","Date":"Mon, 30 Oct 2017 12:21:02 +0100","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.3.0","MIME-Version":"1.0","In-Reply-To":"","Content-Type":"text/plain; charset=utf-8; format=flowed","Content-Transfer-Encoding":"7bit"}},{"id":1800071,"web_url":"http://patchwork.ozlabs.org/comment/1800071/","msgid":"<505927862.453253.1510007773186@poczta.nazwa.pl>","list_archive_url":null,"date":"2017-11-06T22:36:13","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":67752,"url":"http://patchwork.ozlabs.org/api/people/67752/","name":"Rafal Luzynski","email":"digitalfreak@lingonborough.com"},"content":"27.10.2017 18:29 Zack Weinberg wrote:\n> [...]\n> I do not see any new tests anywhere in this patch series, and I found\n> a bug in the support for %OB in strptime (see below). To test this\n> feature we need at least one locale that actually uses it, so rather\n> than gate *these* patches on the addition of tests, I am asking you to\n> write comprehensive tests and include them with the first patch you\n> submit that adds altmon names to a locale. (This locale will have to\n> be one of the locales currently listed in localedata/Makefile as\n> available to tests (the test-input variable, I think); if none of them\n> need this feature, you may add one.)\n\nI've browsed some test and I've found that a similar test already\nexists: time/tst-strptime.c. [1] All we need is to add some more tests\nfor some more locales. Of course we need to add more LOCALES to the\ncorresponding Makefile because now it contains only German, US English,\nand Japanese. I think I should add more than one locale. Please confirm\nthat I'm thinking correctly.\n\nActually we don't need the updated locale data containing the actual\nalternative month names. In most cases the %B/%OB format specifiers\nshould work correctly in strptime() no matter if the input string contains\nthe nominative or genitive case because the algorithm searches for the\nbest match rather than for the equal string.\n\nA longer explanation: The algorithm selects the month name which has\nthe longest matching initial substring with the input string, including\nthe terminating '\\0' character. Since in eastern European languages the\ngrammatical cases are made by appending or changing suffixes while the\nstems (usually) remain the same, this algorithm is still able to\nrecognize the month name correctly. Though, there are cases where\nthis will not work:\n\n* Romance languages using prepositions e.g., Catalan \"de novembre\"\n and \"d’octubre\" will not be recognized correctly; their best match\n without the preposition will be always \"desembre\" (December)\n which is incorrect.\n* Russian and Belarusian \"мая\" (of May) will match both \"май\" (May)\n and \"март\" (March), therefore it will be ambiguous and (if I think\n correctly) it will be recognized as March which is incorrect.\n\nOf course it's worth to add such tests once the genitive month names\nare actually added.\n\n>\n> [...]\n> A note on writing ChangeLog entries: we really do want you to list\n> every single new symbol that's part of the public API, not an\n> abbreviated list -- that is:\n>\n> * locale/langinfo.h [__USE_GNU]: New public symbols ALTMON_1,\n> ALTMON_2, ALTMON_3, ALTMON_4, ALTMON_5, ALTMON_6, ALTMON_7,\n> ALTMON_8, ALTMON_9, ALTMON_10, ALTMON_11, ALTMON_12,\n> _NL_WALTMON_1, _NL_WALTMON_2, _NL_WALTMON_3, _NL_WALTMON_4,\n> _NL_WALTMON_5, _NL_WALTMON_6, _NL_WALTMON_7, _NL_WALTMON_8,\n> _NL_WALTMON_9, _NL_WALTMON_10, _NL_WALTMON_11, _NL_WALTMON_12.\n\nApplied locally. Although I have some doubts. Only ALTMON_x\nsymbols are defined conditionally (#ifdef __USE_GNU) while\n_NL_WALTMON_x are defined unconditionally. Additionally, there\nare __ALTMON_x symbols, also unconditional, defined only because\nALTMON_x are not (yet) defined by POSIX. [2] [3] Should the\nChangeLog be more detailed?\n\n> [ cut longer explanation about ChangeLog ]\n\nAnother longer explanation which I find worth copying to wiki,\nagain if you don't mind, Zack.\n\n> [...]\n> Another minor procedural thing: don't include diffs to the file\n> ChangeLog in commits sent for review, because if a reviewer wants to\n> apply the patch and tinker with it, they don't want to have to clean\n> up the inevitable merge botch in the ChangeLog.\n\nOK, again worth copying to wiki.\n\n> > @@ -402,6 +404,20 @@ __strptime_internal (const char *rp, const char *fmt,\n> > struct tm *tmp,\n> > if (s.decided !=raw)\n> > {\n> > trp = rp;\n> > +#ifdef _LIBC\n> > + /* First check the alt month. */\n>\n> Here you are checking the alt-month name before the month name ...\n>\n> > @@ -428,6 +444,10 @@ __strptime_internal (const char *rp, const char *fmt,\n> > struct tm *tmp,\n> > if (s.decided != loc\n> > && (((trp = rp, match_string (month_name[cnt], trp))\n> > && trp > rp_longest)\n> > +#ifdef _LIBC\n> > + || ((trp = rp, match_string (alt_month_name[cnt], trp))\n> > + && trp > rp_longest)\n> > +#endif\n> > || ((trp = rp, match_string (ab_month_name[cnt], trp))\n> > && trp > rp_longest)))\n>\n> ... and here you are checking the alt-month name after the month name.\n> Please make this consistent one way or the other. (I *think* the\n> order of checks would only ever matter if the alt-month name for month\n> X were the same as the month name for month Y, which would be a nasty\n> ambiguity in the relevant natural language -- but nasty ambiguities do\n> actually happen in natural languages, so we need to be careful.)\n\nIndeed, it should not matter because the algorithm searches for the\nbest matching string which should be only one. If there are two best\nmatching strings then the input string is ambiguous and cannot be\ndetermined, probably the reason is that the input string is wrong.\nIf a language has a string which means two different months depending\non the context then the native speakers have a trouble understanding it.\nThat means, it's unlikely that a language has such a feature. But\nI agree, \"unlikely\" is not \"impossible\". Please note that so far\nI've identified only about 20 languages (out of about 200 supported\nby glibc) which need the nominative/genitive difference in month\nnames, they are in my github repo. [4] It is possible to verify them\nmanually and I think that I did it in the past although I'm not\nsure now.\n\nBut I will apply your remark and introduce the consistency to make\nthe code more readable. So far I was focused on making the *patches*\nmore readable.\n\nNote that the question \"what should be matched first: the basic or\nthe regular month name?\" does not have a good answer. The real good\nanswer should be \"alternative first if the O modifier is present,\nbasic first otherwise\" but since the algorithm drops the modifier\nand treats all specifiers: B, b, and h identically it is difficult\nto tell. Is it worth to store this information and provide different\npaths for the specifiers with the O modifier and without it? I don't\nthink so. Please, note above, most probably that case does not\nactually exist.\n\nWhile at this, I'm somehow concerned by the waste of the CPU cycles\nto check both basic and alternative month names while in 90% languages\nthese strings will always be the same. But I think that checking\nwhether the current locale actually requires the alternative month\nnames and providing separate paths for them would be only worse.\n\n> I wonder whether we actually need the #ifdef _LIBC here -- but that's\n> an independent question (is this code _really_ still shared with\n> gnulib, or have they diverged to the point where we should stop trying\n> to keep the files in sync?) and you are not on the hook to answer it.\n\nOK, I'm unable to answer this reliably but I think that gnulib should\nfollow this change as well. The reason is that the command line utility\ndate uses gnulib rather than glibc. So if we implement the change in\nglibc but do not implement it in other libraries and programs the change\nwill be incomplete.\n\n> > @@ -1015,6 +1035,10 @@ __strptime_internal (const char *rp, const char *fmt,\n> > struct tm *tmp,\n> > case 'O':\n> > switch (*fmt++)\n> > {\n> > + case 'B':\n> > + /* Undo the increment and continue. */\n> > + fmt--;\n> > + break;\n>\n> This is subtly wrong, and I had to read a big chunk of the entire\n> function to understand what it was meant to do and how it doesn't\n> actually work.\n\nOops... :-(\n\nThank you for spotting, analyzing and fixing this.\n\n> Please change to\n>\n> ] case 'O':\n> ] switch (*fmt++)\n> ] {\n> ] + case 'B':\n> ] + /* Match month name. Reprocess as plain 'B'. */\n> ] + fmt--;\n> ] + goto start_over;\n>\n> You will also need to remove the `#ifndef _NL_CURRENT` wrapping the\n> definition of the `start_over` label, and change its comment:\n>\n> ] -#ifndef _NL_CURRENT\n> ] - /* We need this for handling the `E' modifier. */\n> ] + /* In some cases, modifiers are handled by adjusting state and\n> ] + then restarting the switch statement below. */\n> ] start_over:\n> ] -#endif\n> ]\n> ] /* Make back up of current processing pointer. */\n\nI will analyze it more thoroughly but at the first sight your fixes\nare correct and do not require any further changes.\n\n> (Am I seriously telling you to use `goto`? Yes, I am. [...]\n\nI'm OK with goto if there are good reasons to use it, especially\nin the core libraries. Most of the time there are better solutions\nbut this time your reasons are good.\n\nThank you again, regards,\n\nRafal\n\n\n[1] https://sourceware.org/git/?p=glibc.git;a=blob;f=time/tst-strptime.c;hb=HEAD\n[2] https://sourceware.org/ml/libc-alpha/2016-10/msg00303.html\n[3] https://sourceware.org/ml/libc-alpha/2016-12/msg01065.html\n[4] https://github.com/rluzynski/glibc","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86839-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"kcIPRtVc\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yW6qJ6wmCz9s0Z\n\tfor ;\n\tTue, 7 Nov 2017 09:36:28 +1100 (AEDT)","(qmail 124225 invoked by alias); 6 Nov 2017 22:36:21 -0000","(qmail 123974 invoked by uid 89); 6 Nov 2017 22:36:20 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=v7AFrJU5lGV/NBIM\n\tmx1TvM1rRb7PIVtbU0nrHz8TeiciOdocrkxoiMibVaGlsDUekHXsfcRDLOugW7LS\n\t28sybeznIhPkvVbga8whjGcTQpllA29dJ+ZnfPfyRbIEv/5Y2NQo9WewCmUdMsPw\n\tLyVgfQ1xOBScRZrHNoptyz77gds=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; s=default; bh=BL8GfkNAW2yHUM3S/b9/Tj\n\tHbRBk=; b=kcIPRtVcYywneCUFRQ/i0aYGxs9mZXv/QdJjwqrRmbRxEYe1/OWAsJ\n\tUKZQJpeV4DxgVXbmk3qj6vj+ZOxWpROfcYo3ugWsqjsDIu61xnp0iORWA+6ZFy7/\n\tZv2kl+SFS+DXSc5WkyEH24n/2Xilg+vuvpnxjyiXWni+Drh98PcWM=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=2.2 required=5.0 tests=AWL, BAYES_00,\n\tBODY_8BITS, GARBLED_BODY,\n\tKAM_LAZY_DOMAIN_SECURITY autolearn=no version=3.3.2 spammy=novembre,\n\tfocused, gate, comprehensive","X-HELO":"aev204.rev.netart.pl","X-Spam-Score":"2.4","Date":"Mon, 6 Nov 2017 23:36:13 +0100 (CET)","From":"Rafal Luzynski ","Reply-To":"Rafal Luzynski ","To":"Zack Weinberg ","Cc":"GNU C Library ","Message-ID":"<505927862.453253.1510007773186@poczta.nazwa.pl>","In-Reply-To":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","MIME-Version":"1.0","Content-Type":"text/plain; charset=UTF-8","Content-Transfer-Encoding":"quoted-printable","X-Originating-Client":"com.openexchange.ox.gui.dhtml"}},{"id":1800084,"web_url":"http://patchwork.ozlabs.org/comment/1800084/","msgid":"<695842672.453400.1510008776863@poczta.nazwa.pl>","list_archive_url":null,"date":"2017-11-06T22:52:56","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":67752,"url":"http://patchwork.ozlabs.org/api/people/67752/","name":"Rafal Luzynski","email":"digitalfreak@lingonborough.com"},"content":"27.10.2017 19:03 Joseph Myers wrote:\n> [...]\n> Also, the POSIX proposal accepted for issue 8 has strptime %Ob and\n> strptime %OB handled the same, despite that POSIX proposal not including\n> strftime %Ob. (This is only an argument for %Ob handling for strptime to\n> go in this patch rather than the later one adding alternative abbreviated\n> months, however.)\n>\n> --\n> Joseph S. Myers\n> joseph@codesourcery.com\n\nThis may be an omission at POSIX side but I also think that the good\nreason why all variants (B, b, h, OB, Ob, Oh) should be treated the same\nis that the strptime() algorithm should be more forgiving and should\nnot throw the errors like \"yes, I recognize correctly that Nov is\na shortcut for November but the format specifier says you were supposed\nto provide a full version therefore I report an error here\" (also swap\nfull/abbreviated, replace with basic/alternative etc.)\n\nRegards,\n\nRafal","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86841-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"cPigjNLN\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yW7BW5mrQz9s74\n\tfor ;\n\tTue, 7 Nov 2017 09:53:07 +1100 (AEDT)","(qmail 90811 invoked by alias); 6 Nov 2017 22:53:01 -0000","(qmail 90790 invoked by uid 89); 6 Nov 2017 22:53:00 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=y6uTj4tIIqq4ti9V\n\tXT1ms3Kh8e0numlN7QHIev4y8Tt4k9h9virtQXfGrY7LuoWtaOQCeWTUPtKE/dNs\n\txy+TTRKAQZBjB866xWYUQ7A1vVcjrlhVG7Vst65qghFXBnc1XOy74zH1r5dKtS9e\n\t7Nj0nhv59UsrA6ijKii9bakJfhs=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; s=default; bh=sgFrkndrPioOs4s+soabUa\n\tsxuHo=; b=cPigjNLNX1wZ6NI9IPTyQYqHsuFps0gvPzuUKQz5j8GO3l1f89ciy5\n\tps9ZmFeuDrREQCi//gHzmuBhrMe/g/WhvN58miDv03M03X5nVM6Ooi30skGmQrPZ\n\trNCr9CjPr3EuPTcmYZz1abup0TOZxAfyo79udFlt8MIhlNLDwl8gI=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.4 required=5.0 tests=AWL, BAYES_00,\n\tKAM_LAZY_DOMAIN_SECURITY,\n\tURIBL_RED autolearn=no version=3.3.2 spammy=HImportance:Medium,\n\tH*x:Open-Xchange, H*UA:Open-Xchange, month","X-HELO":"aev204.rev.netart.pl","X-Spam-Score":"0","Date":"Mon, 6 Nov 2017 23:52:56 +0100 (CET)","From":"Rafal Luzynski ","Reply-To":"Rafal Luzynski ","To":"Zack Weinberg , Joseph Myers ","Cc":"GNU C Library ","Message-ID":"<695842672.453400.1510008776863@poczta.nazwa.pl>","In-Reply-To":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","MIME-Version":"1.0","Content-Type":"text/plain; charset=UTF-8","Content-Transfer-Encoding":"7bit","X-Originating-Client":"com.openexchange.ox.gui.dhtml"}},{"id":1800088,"web_url":"http://patchwork.ozlabs.org/comment/1800088/","msgid":"<831445356.453465.1510009278439@poczta.nazwa.pl>","list_archive_url":null,"date":"2017-11-06T23:01:18","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":67752,"url":"http://patchwork.ozlabs.org/api/people/67752/","name":"Rafal Luzynski","email":"digitalfreak@lingonborough.com"},"content":"27.10.2017 22:16 Zack Weinberg wrote:\n>\n>\n> On Fri, Oct 27, 2017 at 3:25 PM, Joseph Myers wrote:\n> > On Fri, 27 Oct 2017, Zack Weinberg wrote:\n> >>> On Fri, Oct 27, 2017 at 1:19 PM, Joseph Myers \n> >>> wrote:\n> >> > http://austingroupbugs.net/view.php?id=258 (issue8 tag = for next major\n> >> > revision).\n> >>\n> >> Hmm. Should we, somehow, tell them about the need for %Ob?\n> >\n> > Yes, probably. I believe anyone can create an account to file or comment\n> > on bugs.\n>\n> I have filed http://austingroupbugs.net/view.php?id=1166 .\n\nThank you Zack for filing this, also thank you Joseph for filing an\nissue against GCC. Your reports are perfect and prove that you understand\nthe problem.\n\nMy only concern is that both reports in austingroupbugs still refer to\nthe actual grammatical cases (genitive and nominative) which already lead\nto some misunderstandings here in this list. It has been stated that\nit's more correct to refer to \"the form correct when the month is used\nto format a full date\" and \"the form correct when the month is named\nby itself\". In some languages the genitive case exists and it's incorrect\nto use it when formatting a full date.\n\nRegards,\n\nRafal","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86843-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"rv4tq/cB\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yW7N81VxNz9sCZ\n\tfor ;\n\tTue, 7 Nov 2017 10:01:27 +1100 (AEDT)","(qmail 50454 invoked by alias); 6 Nov 2017 23:01:22 -0000","(qmail 50443 invoked by uid 89); 6 Nov 2017 23:01:22 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=rHC00ZZKeFHPaYUS\n\thV7dkuGeHc6pjMZTXwBHQ3Eh7pVG892CZG0hwt8hb6Cq/iRGIvuWeQMR89sV/TRF\n\tcSmfnieppdJHcayVmzaT+v9RtGJLpmbB/tG54k6F7vN2zoecFlTpN7plbPkGneCl\n\t0xWGPajqV60CIGc++LmDHx8eSf8=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; s=default; bh=RML4BdSBIbY5Yuhex2yH36\n\tfS+4M=; b=rv4tq/cB03ZHJWtue0Y2vFGtgMeCjvsJn4O0cQBN5sUNimnTBS+CLo\n\tuWRP0gLzP+w8smtR6HPqN+3XH4k2V5VA0H9Tc8yE5prXSVjtMFaclZ/W7bJj9a8d\n\tECjPFnuj8Y1/Dvlh9eiS/UI1KoMe8el+vzejWk0xGciLtj9UM+xX4=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.4 required=5.0 tests=AWL, BAYES_00,\n\tKAM_LAZY_DOMAIN_SECURITY,\n\tURIBL_RED autolearn=no version=3.3.2 spammy=filed,\n\tHImportance:Medium, H*x:Open-Xchange, filing","X-HELO":"aev204.rev.netart.pl","X-Spam-Score":"0.8","Date":"Tue, 7 Nov 2017 00:01:18 +0100 (CET)","From":"Rafal Luzynski ","Reply-To":"Rafal Luzynski ","To":"Zack Weinberg , Joseph Myers ","Cc":"GNU C Library ","Message-ID":"<831445356.453465.1510009278439@poczta.nazwa.pl>","In-Reply-To":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t\n\t\n\t\n\t\n\t\n\t","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","MIME-Version":"1.0","Content-Type":"text/plain; charset=UTF-8","Content-Transfer-Encoding":"7bit","X-Originating-Client":"com.openexchange.ox.gui.dhtml"}},{"id":1800091,"web_url":"http://patchwork.ozlabs.org/comment/1800091/","msgid":"<675837978.453531.1510009738323@poczta.nazwa.pl>","list_archive_url":null,"date":"2017-11-06T23:08:58","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":67752,"url":"http://patchwork.ozlabs.org/api/people/67752/","name":"Rafal Luzynski","email":"digitalfreak@lingonborough.com"},"content":"27.10.2017 22:36 Zack Weinberg wrote:\n> On Fri, Oct 27, 2017 at 4:30 PM, Joseph Myers wrote:\n> > On Fri, 27 Oct 2017, Zack Weinberg wrote:\n> >> [...]\n> >> I have filed http://austingroupbugs.net/view.php?id=1166 .\n> >\n> > I note that includes ABALTMON_* constants in langinfo.h, whereas the\n> > present patch series has ALTMON_* constants but no ABALTMON_* names.\n>\n> Rafal's patches have _NL_ABALTMON_*. I am not clear on when we\n> actually want _NL_ prefixes on those constants, but I figured it was\n> simpler to write the proposal with no prefixes and the Austin Group\n> can add them if they want them.\n\nIf I understand correctly, _NL_* prefixes are used for our own extensions\nwhile the constants without the prefixes are standardized by POSIX.\n\nI can't define ABALTMON_* constants before POSIX accepts and publishes\nthe change. Should I define them now when the issue is filed? Can\nI assume that if the change is filed it will be eventually accepted\njust because it looks reasonable for us? I don't think so. It may take\nmultiple years. How should we prepare for the potential change in the\ndistant future? Is it OK to leave it as it is now and assume that\nABALTMON_* symbols will be defined (as aliases) when they are needed?\n\nRegards,\n\nRafal","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-86844-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"UzgMvPxS\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yW7Y15HK9z9sCZ\n\tfor ;\n\tTue, 7 Nov 2017 10:09:09 +1100 (AEDT)","(qmail 61028 invoked by alias); 6 Nov 2017 23:09:02 -0000","(qmail 61010 invoked by uid 89); 6 Nov 2017 23:09:02 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=VmeXdJtcyysAQjLT\n\tAPOLmL6GllxaaArLv4Ukbpk8HSQQz0JkVTdC+OqETN+RCnnXGrynHW5ZKx7gRttr\n\tigszp9jLR+HrlWBJgWaiwfmk8fy00DnZyiI7hku9lozApthNMP8J+/w27civNO51\n\tGXDZNSTdqTrO6WQChB2RsvjZddc=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; s=default; bh=+2ZWyPEf9oxayE5oc1PXTW\n\tXeVhk=; b=UzgMvPxSrf98l68v++vGsVfmlZy+CWNeKF+Nd/x/ghO6GQpBlid+dB\n\tJfoBkpLYvvZJ9h4nXrBMSiLSketFM5tnDAui+WEYJ0AenlDGuErUvu1R4waD5FKD\n\tdya2KdulKitxRnhvmWeYjYgMyOiRsQXwjjQoZvUCVK7vjFIAqaYLo=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.4 required=5.0 tests=AWL, BAYES_00,\n\tKAM_LAZY_DOMAIN_SECURITY,\n\tURIBL_RED autolearn=no version=3.3.2 spammy=Austin, austin,\n\tfigured, filed","X-HELO":"aev204.rev.netart.pl","X-Spam-Score":"2.8","Date":"Tue, 7 Nov 2017 00:08:58 +0100 (CET)","From":"Rafal Luzynski ","Reply-To":"Rafal Luzynski ","To":"Zack Weinberg , Joseph Myers ","Cc":"GNU C Library ","Message-ID":"<675837978.453531.1510009738323@poczta.nazwa.pl>","In-Reply-To":"","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t\n\t\n\t\n\t\n\t\n\t\n\t\n\t","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","MIME-Version":"1.0","Content-Type":"text/plain; charset=UTF-8","Content-Transfer-Encoding":"7bit","X-Originating-Client":"com.openexchange.ox.gui.dhtml"}},{"id":1804918,"web_url":"http://patchwork.ozlabs.org/comment/1804918/","msgid":"<283973500.20891.1510741959500@poczta.nazwa.pl>","list_archive_url":null,"date":"2017-11-15T10:32:38","subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","submitter":{"id":67752,"url":"http://patchwork.ozlabs.org/api/people/67752/","name":"Rafal Luzynski","email":"digitalfreak@lingonborough.com"},"content":"6.11.2017 23:36 Rafal Luzynski wrote:\n> [...]\n> Actually we don't need the updated locale data containing the actual\n> alternative month names. In most cases the %B/%OB format specifiers\n> should work correctly in strptime() no matter if the input string contains\n> the nominative or genitive case because the algorithm searches for the\n> best match rather than for the equal string.\n>\n> A longer explanation: The algorithm selects the month name which has\n> the longest matching initial substring with the input string, including\n> the terminating '\\0' character. Since in eastern European languages the\n> grammatical cases are made by appending or changing suffixes while the\n> stems (usually) remain the same, this algorithm is still able to\n> recognize the month name correctly. [...]\n\nI made some tests locally and it turns out I was wrong here. The\ngenitive forms (in eastern European languages) will not be recognized\nby the current locales just because they look similar to the nominative\nforms. For example, in Polish language (I choose the simplest word)\nthe word for May is \"maj\", the genitive form is \"maja\". When we try\nto parse \"maja\" with strptime(\"%B\", ...) we get the substring \"maj\"\ncorrectly recognized as the 5th month while the letter \"a\" remains\nunparsed which eventually raises an error. I incorrectly thought\nthat strptime() matches the whole word and returns the index of the\nword from the repository (from nl_langinfo() results) which has the\nlongest matching substring (whole string is the best). Actually\nit matches the longest substring and leaves the rest of the word\nunparsed.\n\nShortly, only the words which are actually present in the repository\nare recognized as the month names, or their initial substrings.\nThe words which partially match raise errors.\n\nOn the other hand, \"%OB\" format specifier should work fine\n(as well as \"%B\") as long as we use only the nominative cases\nin the input strings.\n\nRegards,\n\nRafal","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-87126-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"NH/+Hup2\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3ycLLD4Zcrz9s7G\n\tfor ;\n\tWed, 15 Nov 2017 21:32:52 +1100 (AEDT)","(qmail 93787 invoked by alias); 15 Nov 2017 10:32:46 -0000","(qmail 93770 invoked by uid 89); 15 Nov 2017 10:32:45 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=RfIqwrRBNlFHtDHt\n\tAP7aOrf4eESpd8vuoQ47q+CGHEKRkqOBD9TJHXQMrQD7+RTjkEuQIUYS4V4gNc5t\n\tT47NtjPq0lBwcvnfWs0dHbaQ23RMIw7JG+4IK3rL+9W0BT9mcAii5uWp30F9DJCr\n\tJJBKyqm914xJbj3BZ39srUuYqTE=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:reply-to:to:cc:message-id\n\t:in-reply-to:references:subject:mime-version:content-type\n\t:content-transfer-encoding; s=default; bh=PwZAKiLCdYlPZY/HhN66Le\n\t7xqYg=; b=NH/+Hup21uBNmHUmlAKmEEv1pIySaxGMjopOIX1SZdKhIMf6G9RF8y\n\tVWkVEZtOIbCLjbtvvsu2c0jj7oyCgIAqxLN72KDsJ9rh9z+sqcF5ph/HvCm6yf+y\n\t+yFtZ12Z5BZXAoEa/Y4N52eZ/JQL06h3pEqTh9GagiWkfAqAMgwgw=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-5.7 required=5.0 tests=AWL, BAYES_00,\n\tGIT_PATCH_1, KAM_LAZY_DOMAIN_SECURITY,\n\tKB_WAM_FROM_NAME_SINGLEWORD autolearn=ham version=3.3.2\n\tspammy=H*r:10.252.0, HTo:D*panix.com, month, letter","X-HELO":"aev204.rev.netart.pl","X-Spam-Score":"0","Date":"Wed, 15 Nov 2017 11:32:38 +0100 (CET)","From":"Rafal Luzynski ","Reply-To":"Rafal Luzynski ","To":"Zack Weinberg ","Cc":"GNU C Library ","Message-ID":"<283973500.20891.1510741959500@poczta.nazwa.pl>","In-Reply-To":"<505927862.453253.1510007773186@poczta.nazwa.pl>","References":"<742475879.1094767.1505817734249@poczta.nazwa.pl>\n\t\n\t<505927862.453253.1510007773186@poczta.nazwa.pl>","Subject":"Re: [RFC][PATCH v9 2/6] Implement alternative month names (bug\n\t10871).","MIME-Version":"1.0","Content-Type":"text/plain; charset=UTF-8","Content-Transfer-Encoding":"7bit","X-Originating-Client":"com.openexchange.ox.gui.dhtml"}}]