izimobil - Project Hosting on Google Code

User Profile		Updates

Last 7 days Dec 18, 2009 django-mptt issue 65 (Allow select_related in model methods) reported - The attached patch allows to pass an extra keyword argument "select_related" to the following methods: - get_ancestors() - get_children() - get_descendants() - get_next_sibling() - get_previous_sibling() - get_siblings() and in the util function 'drilldown_tree_for_node()'. It can reduce the number of queries in many usecases (eg: breadcrumbs, menus, etc...). The attached patch allows to pass an extra keyword argument "select_related" to the following methods: - get_ancestors() - get_children() - get_descendants() - get_next_sibling() - get_previous_sibling() - get_siblings() and in the util function 'drilldown_tree_for_node()'. It can reduce the number of queries in many usecases (eg: breadcrumbs, menus, etc...). Dec 17, 2009 onlogistics r399 (merge with trunk) committed - merge with trunk merge with trunk Dec 17, 2009 onlogistics r398 (merge with trunk) committed - merge with trunk merge with trunk Dec 17, 2009 onlogistics r397 (actor.code was not displayed) committed - actor.code was not displayed actor.code was not displayed Dec 14, 2009 polib issue 25 (setup.py requires CHANGELOG, but it's not present in the pol...) changed - Argh :( I've just released a 0.5.1 that fixes the issue, thanks diego ! Status: Fixed Owner: izimobil Argh :( I've just released a 0.5.1 that fixes the issue, thanks diego ! Status: Fixed Owner: izimobil Dec 14, 2009 polib polib-0.5.1.tar.gz (polib version 0.5.1) file uploaded - Labels: Featured Type-Source OpSys-All Labels: Featured Type-Source OpSys-All Dec 14, 2009 polib Revision 0940388b67 (- fixed issue #0025: setup.py requires CHANGELOG but it's no...) pushed - - fixed issue #0025: setup.py requires CHANGELOG but it's not present in polib-0.5.0-tar.gz - new release 0.5.1 - fixed issue #0025: setup.py requires CHANGELOG but it's not present in polib-0.5.0-tar.gz - new release 0.5.1 Dec 13, 2009 polib polib-0.5.0.tar.gz (polib version 0.5.0) file uploaded - Labels: Featured Type-Source OpSys-All Labels: Featured Type-Source OpSys-All Dec 13, 2009 polib polib-0.5.0.tar.gz (polib version 0.5.0) file uploaded - Labels: Featured Type-Source OpSys-All Labels: Featured Type-Source OpSys-All Dec 13, 2009 polib Revision af63555a8b (oups, removed print statement) pushed - oups, removed print statement oups, removed print statement Dec 13, 2009 polib Revision 6b65dc62b0 (rest formatting + simplify deployment to the holy cheeseshop) pushed - rest formatting + simplify deployment to the holy cheeseshop rest formatting + simplify deployment to the holy cheeseshop Last 30 days Dec 13, 2009 polib polib-0.5.0.tar.gz (polib version 0.5.0) file uploaded - Labels: Featured Type-Source OpSys-All Labels: Featured Type-Source OpSys-All Dec 13, 2009 polib Revision d8be9f5e3a (finally, it will be: release 0.5.0) pushed - finally, it will be: release 0.5.0 finally, it will be: release 0.5.0 Dec 13, 2009 polib issue 23 (Encoding error when saving mo file as po file) changed - Hi Sebastien, I applied your patch, thanks ! http://code.google.com/p/polib/source/detail?r=d285dc7160372d7a81b03db2da375ea9619d911f -- David Status: Fixed Owner: izimobil Hi Sebastien, I applied your patch, thanks ! http://code.google.com/p/polib/source/detail?r=d285dc7160372d7a81b03db2da375ea9619d911f -- David Status: Fixed Owner: izimobil Dec 13, 2009 polib Revision d285dc7160 (* parse msg plural entries correctly when reading mo files, ...) pushed - * parse msg plural entries correctly when reading mo files, * fixed issue #23: encoding error when saving mo file as po file (thanks to sebastien.sable for the patch !). * parse msg plural entries correctly when reading mo files, * fixed issue #23: encoding error when saving mo file as po file (thanks to sebastien.sable for the patch !). Dec 11, 2009 polib ProjectsUsingPolib Wiki page edited Dec 11, 2009 polib issue 24 (Plan to release 0.4.3 with msgctxt support) changed - Hi, please do not open an issue just to ask a question. I'll release 0.4.3 very soon, as soon as I get a msg plural bug fixed. oh btw, may I add Transifex here ? http://code.google.com/p/polib/wiki/ProjectsUsingPolib -- David Status: Invalid Owner: izimobil Hi, please do not open an issue just to ask a question. I'll release 0.4.3 very soon, as soon as I get a msg plural bug fixed. oh btw, may I add Transifex here ? http://code.google.com/p/polib/wiki/ProjectsUsingPolib -- David Status: Invalid Owner: izimobil Dec 02, 2009 onlogistics r396 (merge with trunk) committed - merge with trunk merge with trunk Dec 02, 2009 onlogistics r395 (bullet proof script when there's no images) committed - bullet proof script when there's no images bullet proof script when there's no images Dec 02, 2009 onlogistics r394 (merge with trunk) committed - merge with trunk merge with trunk Dec 02, 2009 onlogistics r393 (save preferences) committed - save preferences save preferences Dec 02, 2009 onlogistics r392 (added EAN13 code + give access to command list to owner cust...) committed - added EAN13 code + give access to command list to owner customer added EAN13 code + give access to command list to owner customer Nov 29, 2009 polib issue 22 (Unescaping code is insufficient) Status changed - Status: Fixed Status: Fixed Nov 29, 2009 polib Revision bfbe8aba8b (fixed issue #22: unescaping code is insufficient) pushed - fixed issue #22: unescaping code is insufficient fixed issue #22: unescaping code is insufficient Nov 27, 2009 polib Tutorial Wiki page edited Nov 27, 2009 polib Readme Wiki page edited Nov 27, 2009 polib Readme Wiki page edited Nov 27, 2009 polib Readme Wiki page edited Nov 27, 2009 polib Readme Wiki page edited Nov 27, 2009 polib Readme Wiki page edited Nov 27, 2009 polib Readme Wiki page edited Nov 27, 2009 polib issue 22 (Unescaping code is insufficient) commented on - > seems to pass the tests. Unfortunately no, and actually I get 32 failures ! Reasons for this are: - string_escape also escapes the quote (') character, we do not want this for po files, - string_escape can't handle unicode strings (I also tried unicode_escape but for some reason it still doesn't work), - string_escape, as you noted, doesn't escape the double quote character (but this can easily get fixed). So it's not that easy... I think I'll apply the patch and keep the functions as is, they pass all tests. Thanks for your report ! -- David > seems to pass the tests. Unfortunately no, and actually I get 32 failures ! Reasons for this are: - string_escape also escapes the quote (') character, we do not want this for po files, - string_escape can't handle unicode strings (I also tried unicode_escape but for some reason it still doesn't work), - string_escape, as you noted, doesn't escape the double quote character (but this can easily get fixed). So it's not that easy... I think I'll apply the patch and keep the functions as is, they pass all tests. Thanks for your report ! -- David Nov 27, 2009 polib issue 22 (Unescaping code is insufficient) Status changed - Hmm now I understand, I attach a fix for this, I'm not sure it's the best thing to do but at least it fixes the bug and don't break existing tests. Can you test the patch and tell me if it's ok for you or if you think you can improve the patch ? Status: Started Hmm now I understand, I attach a fix for this, I'm not sure it's the best thing to do but at least it fixes the bug and don't break existing tests. Can you test the patch and tell me if it's ok for you or if you think you can improve the patch ? Status: Started Nov 26, 2009 polib ProjectsUsingPolib Wiki page added Nov 26, 2009 polib ProjectsUsingPolib2 Wiki page deleted Nov 26, 2009 polib Tutorial Wiki page added Nov 26, 2009 polib Readme Wiki page added Nov 26, 2009 polib ProjectsUsingPolib2 (One-sentence summary of this page.) Wiki page edited Nov 26, 2009 polib ProjectsUsingPolib2 (One-sentence summary of this page.) Wiki page added Nov 26, 2009 polib issue 22 (Unescaping code is insufficient) commented on - (I'll try to quote your previous messages, but it's not easy with googlecode issue tracker...) > This seems to be because unescape(), when faced with the escape sequence \\n, > chooses to first replace \n with a newline, rather than recognizing that the > double backslash is intended to prevent this from being considered a newline. Well, that's the point of "unescaping" a string, I mean: escape "\n" (newline) produces "\\n" and unescape "\\n" produces "\n" (newline), that's the correct reverse operation. > We cann see that unescape() doesn't reverse the effects of escape() in this case, > but mangles the string. that's because you "print" the result and your terminal interprets the newline char, again, to me that's correct: >>> from polib import escape, unescape >>> escape("\n") '\\n' >>> unescape("\\n") '\n' > This would potentially mean that a sequence like \ + n doesn't need to be escaped > at all inside a pofile, and that xml2po writes out the wrong strings Correct. I think the problem is on xml2po script, it should not escape the "\n" in your xml file, in a po file, a newline is represented by "\n", look at some django po file for example, you'll see that "\n" are not escaped in the file. So given this xml string: <string name="button_caption">Starŧ\nNow</string> xml2po should produce: msgid "Start\nNow" > However, I don't think it would change the fact that polib still unescapes a > string like "\\n" incorrectly. polib does the correct thing I think, unescape("\\n") produces "\n" and if you print it, you've got a newline: >>> print "\\n" \n >>> print escape("\n") \n >>> print "\n" >>> print unescape("\\n") -- David (I'll try to quote your previous messages, but it's not easy with googlecode issue tracker...) > This seems to be because unescape(), when faced with the escape sequence \\n, > chooses to first replace \n with a newline, rather than recognizing that the > double backslash is intended to prevent this from being considered a newline. Well, that's the point of "unescaping" a string, I mean: escape "\n" (newline) produces "\\n" and unescape "\\n" produces "\n" (newline), that's the correct reverse operation. > We cann see that unescape() doesn't reverse the effects of escape() in this case, > but mangles the string. that's because you "print" the result and your terminal interprets the newline char, again, to me that's correct: >>> from polib import escape, unescape >>> escape("\n") '\\n' >>> unescape("\\n") '\n' > This would potentially mean that a sequence like \ + n doesn't need to be escaped > at all inside a pofile, and that xml2po writes out the wrong strings Correct. I think the problem is on xml2po script, it should not escape the "\n" in your xml file, in a po file, a newline is represented by "\n", look at some django po file for example, you'll see that "\n" are not escaped in the file. So given this xml string: <string name="button_caption">Starŧ\nNow</string> xml2po should produce: msgid "Start\nNow" > However, I don't think it would change the fact that polib still unescapes a > string like "\\n" incorrectly. polib does the correct thing I think, unescape("\\n") produces "\n" and if you print it, you've got a newline: >>> print "\\n" \n >>> print escape("\n") \n >>> print "\n" >>> print unescape("\\n") -- David Nov 26, 2009 polib issue 22 (Unescaping code is insufficient) commented on - Sorry but to me that's correct: >>> import polib >>> r'\\n' '\\\\n' >>> polib.unescape(r'\\n') '\\\n' >>> polib.unescape('\\\\n') '\\\n' Polib has been tested against many real world po files and this problem never occured, can you give me your usecase ? are you reading an existing pofile ? creating a new one from scratch ? Thanks, -- David Sorry but to me that's correct: >>> import polib >>> r'\\n' '\\\\n' >>> polib.unescape(r'\\n') '\\\n' >>> polib.unescape('\\\\n') '\\\n' Polib has been tested against many real world po files and this problem never occured, can you give me your usecase ? are you reading an existing pofile ? creating a new one from scratch ? Thanks, -- David Nov 26, 2009 polib issue 22 (Unescaping code is insufficient) commented on - I'm not sure I understand the issue. Please can you provide a sample of code with what you expect and what polib does ? Thanks, -- David I'm not sure I understand the issue. Please can you provide a sample of code with what you expect and what polib does ? Thanks, -- David Earlier this year Oct 31, 2009 polib issue 21 ('append' method might duplicate entries) changed - I have implemented the feature request, now to force polib to check for duplicate you just have to do: >>> import polib >>> po = polib.pofile('test.po', check_for_duplicates=True) It also works for mo files: >>> import polib >>> po = polib.mofile('test.mo', check_for_duplicates=True) And also when using the POFile/MOFile classes directly: >>> import polib >>> po = polib.POFile(check_for_duplicates=True) >>> po.append(polib.POEntry(msgid='foo')) >>> po.append(polib.POEntry(msgid='bar')) >>> po.append(polib.POEntry(msgid='foo')) Traceback (most recent call last): File "<stdin>", line 1, in <module> File "polib.py", line 349, in append raise ValueError('Entry "%s" already exists' % entry.msgid) ValueError: Entry "foo" already exists This behavior is intentionaly optional, because it really slows down things (especially with big catalogs). -- David Status: Fixed Owner: izimobil I have implemented the feature request, now to force polib to check for duplicate you just have to do: >>> import polib >>> po = polib.pofile('test.po', check_for_duplicates=True) It also works for mo files: >>> import polib >>> po = polib.mofile('test.mo', check_for_duplicates=True) And also when using the POFile/MOFile classes directly: >>> import polib >>> po = polib.POFile(check_for_duplicates=True) >>> po.append(polib.POEntry(msgid='foo')) >>> po.append(polib.POEntry(msgid='bar')) >>> po.append(polib.POEntry(msgid='foo')) Traceback (most recent call last): File "<stdin>", line 1, in <module> File "polib.py", line 349, in append raise ValueError('Entry "%s" already exists' % entry.msgid) ValueError: Entry "foo" already exists This behavior is intentionaly optional, because it really slows down things (especially with big catalogs). -- David Status: Fixed Owner: izimobil Oct 31, 2009 polib Revision 90c1b39188 (- Some cleanup, - Added ability to check for duplicate when ...) pushed - - Some cleanup, - Added ability to check for duplicate when adding entries to po/mo files, this is optional and not enabled by default because it slows down considerably the library. - Some cleanup, - Added ability to check for duplicate when adding entries to po/mo files, this is optional and not enabled by default because it slows down considerably the library. Oct 30, 2009 polib issue 20 (Duplicated entries) commented on - Yeah, the entry is modified by reference. The "already exists" exception on append is a good idea, can you open a separated feature request for this please ? I'll implement it ASAP. Thanks. Yeah, the entry is modified by reference. The "already exists" exception on append is a good idea, can you open a separated feature request for this please ? I'll implement it ASAP. Thanks. Oct 30, 2009 polib issue 20 (Duplicated entries) Status changed - Hi, this is not a bug, you explicitely "append" again the entry in the pofile, so obviously polib add another entry, to achieve what you want, just remove this line: In [7]: po.append(poentry) Status: Invalid Hi, this is not a bug, you explicitely "append" again the entry in the pofile, so obviously polib add another entry, to achieve what you want, just remove this line: In [7]: po.append(poentry) Status: Invalid Oct 20, 2009 polib Revision 6de03ed98f (prepare for release 0.4.3) pushed - prepare for release 0.4.3 prepare for release 0.4.3 Oct 20, 2009 polib 2 new revisions pushed - a0a8e70ee2:removed API docs 5da4af4ea2:prepare for release 0.4.3 a0a8e70ee2:removed API docs 5da4af4ea2:prepare for release 0.4.3 Oct 20, 2009 polib Revision 7f5a7dd5ad (fixed bug when compiling plural msgids/strs) pushed - fixed bug when compiling plural msgids/strs fixed bug when compiling plural msgids/strs Oct 20, 2009 polib issue 19 (Latest polib breaks plural forms -- "not NUL terminated stri...) Status changed - Nevermind, this should be fixed now. -- David Status: Fixed Nevermind, this should be fixed now. -- David Status: Fixed

issue 22 (Unescaping code is insufficient) commented on - (I'll try to quote your previous messages, but it's not easy with googlecode issue tracker...) > This seems to be because unescape(), when faced with the escape sequence \\n, > chooses to first replace \n with a newline, rather than recognizing that the > double backslash is intended to *prevent* this from being considered a newline. Well, that's the point of "unescaping" a string, I mean: escape "\n" (newline) produces "\\n" and unescape "\\n" produces "\n" (newline), that's the correct reverse operation. > We cann see that unescape() doesn't reverse the effects of escape() in this case, > but mangles the string. that's because you "print" the result and your terminal interprets the newline char, again, to me that's correct: >>> from polib import escape, unescape >>> escape("\n") '\\n' >>> unescape("\\n") '\n' > This would potentially mean that a sequence like \ + n doesn't need to be escaped > at all inside a pofile, and that xml2po writes out the wrong strings Correct. I think the problem is on xml2po script, it should *not* escape the "\n" in your xml file, in a po file, a newline is represented by "\n", look at some django po file for example, you'll see that "\n" are not escaped in the file. So given this xml string: <string name="button_caption">Starŧ\nNow</string> xml2po should produce: msgid "Start\nNow" > However, I don't think it would change the fact that polib still unescapes a > string like "\\n" incorrectly. polib does the correct thing I think, unescape("\\n") produces "\n" and if you print it, you've got a newline: >>> print "\\n" \n >>> print escape("\n") \n >>> print "\n" >>> print unescape("\\n") -- David

(I'll try to quote your previous messages, but it's not easy with googlecode issue tracker...) > This seems to be because unescape(), when faced with the escape sequence \\n, > chooses to first replace \n with a newline, rather than recognizing that the > double backslash is intended to *prevent* this from being considered a newline. Well, that's the point of "unescaping" a string, I mean: escape "\n" (newline) produces "\\n" and unescape "\\n" produces "\n" (newline), that's the correct reverse operation. > We cann see that unescape() doesn't reverse the effects of escape() in this case, > but mangles the string. that's because you "print" the result and your terminal interprets the newline char, again, to me that's correct: >>> from polib import escape, unescape >>> escape("\n") '\\n' >>> unescape("\\n") '\n' > This would potentially mean that a sequence like \ + n doesn't need to be escaped > at all inside a pofile, and that xml2po writes out the wrong strings Correct. I think the problem is on xml2po script, it should *not* escape the "\n" in your xml file, in a po file, a newline is represented by "\n", look at some django po file for example, you'll see that "\n" are not escaped in the file. So given this xml string: <string name="button_caption">Starŧ\nNow</string> xml2po should produce: msgid "Start\nNow" > However, I don't think it would change the fact that polib still unescapes a > string like "\\n" incorrectly. polib does the correct thing I think, unescape("\\n") produces "\n" and if you print it, you've got a newline: >>> print "\\n" \n >>> print escape("\n") \n >>> print "\n" >>> print unescape("\\n") -- David

Last 7 days

Last 30 days

Earlier this year