From: Kees B. <kee...@xs...> - 2006-11-08 19:31:28
|
Hi, This is of interest for all translators. A new version check_po has been checked-in. It is a rewrite of the script but with most of the functionallity in place. Most obvious change is that the format checks have been improved, so it will catch more potential errors. And also there is a new check which looks for XML special characters in the translated strings. Some characters must be given in their XML format. Examples are: double quote, ampersand, angle bracket. My plan is to improve it a little bit further. For example, it flags an error if there are percent signs in the msgstr but not in the msgid. This is not a fatal error, and I want to separate them in their own list. If you have problems with the new check_po, please let me know. - Kees |
From: Alex R. <sh...@gr...> - 2006-11-08 20:57:53
|
Kees, The new script works great for me, thanks! Once question for you: when it prints the problem strings, what is the first number reported? Is that the physical line in the po file, or a msgid/msgstr number? Alex On Wed, 2006-11-08 at 20:30 +0100, Kees Bakker wrote: > This is of interest for all translators. A new version check_po > has been checked-in. It is a rewrite of the script but with > most of the functionallity in place. >=20 > Most obvious change is that the format checks have been improved, so > it will catch more potential errors. >=20 > And also there is a new check which looks for XML special characters > in the translated strings. Some characters must be given in their > XML format. Examples are: double quote, ampersand, angle bracket. >=20 > My plan is to improve it a little bit further. For example, it > flags an error if there are percent signs in the msgstr but not in > the msgid. This is not a fatal error, and I want to separate them > in their own list. >=20 > If you have problems with the new check_po, please let me know. --=20 Alexander Roitman http://www.gramps-project.org |
From: Kees B. <kee...@xs...> - 2006-11-09 07:08:56
|
Hi Alex, Yes, that number is the linenumber. It sometimes comes in handy when you want to find that spot. - Kees Op woensdag 08 november 2006 21:57, schreef Alex Roitman: > Kees, > > The new script works great for me, thanks! > > Once question for you: when it prints the problem strings, > what is the first number reported? Is that the physical line > in the po file, or a msgid/msgstr number? > > Alex > > On Wed, 2006-11-08 at 20:30 +0100, Kees Bakker wrote: > > This is of interest for all translators. A new version check_po > > has been checked-in. It is a rewrite of the script but with > > most of the functionallity in place. > > > > Most obvious change is that the format checks have been improved, so > > it will catch more potential errors. > > > > And also there is a new check which looks for XML special characters > > in the translated strings. Some characters must be given in their > > XML format. Examples are: double quote, ampersand, angle bracket. > > > > My plan is to improve it a little bit further. For example, it > > flags an error if there are percent signs in the msgstr but not in > > the msgid. This is not a fatal error, and I want to separate them > > in their own list. > > > > If you have problems with the new check_po, please let me know. > |
From: Alex R. <sh...@gr...> - 2006-11-09 07:17:42
|
Kees, On Thu, 2006-11-09 at 08:08 +0100, Kees Bakker wrote: > Yes, that number is the linenumber. It sometimes comes in handy when > you want to find that spot. Thanks, this makes sense. What do you think about two suggestions for the script: 1. Also report the msgid numbers: e.g. kbabel lists the numbers of msgid/msgstr and it's easy to find the needed strings by that number. 2. Add a newline between the listed problematic lines. At present it's tough to see what's what because everything is mushed together. Not big problems, just see if this makes sense for you. Thanks, Alex --=20 Alexander Roitman http://www.gramps-project.org |
From: Eero T. <ee...@us...> - 2006-11-10 23:09:59
|
Hi, On Thursday 09 November 2006 09:17, Alex Roitman wrote: > > Yes, that number is the linenumber. It sometimes comes in handy when > > you want to find that spot. > > Thanks, this makes sense. > > What do you think about two suggestions for the script: > > 1. Also report the msgid numbers: e.g. kbabel lists the > numbers of msgid/msgstr and it's easy to find the needed > strings by that number. > > 2. Add a newline between the listed problematic lines. > At present it's tough to see what's what because everything > is mushed together. I see these are now added and it indeed makes the output clearer. I currently have a lot of stuff to translate, maybe the script could also show which of the incorrect strings are fuzzy, e.g. like this: msg nr: 1487, lineno: 7704 (fuzzy) msgid "Summary" msgstr "%s; yhteenveto" ? - Eero |
From: Kees B. <kee...@xs...> - 2006-11-11 08:29:11
|
Op zaterdag 11 november 2006 00:15, schreef Eero Tamminen: > Hi, > > On Thursday 09 November 2006 09:17, Alex Roitman wrote: > > > Yes, that number is the linenumber. It sometimes comes in handy when > > > you want to find that spot. > > > > Thanks, this makes sense. > > > > What do you think about two suggestions for the script: > > > > 1. Also report the msgid numbers: e.g. kbabel lists the > > numbers of msgid/msgstr and it's easy to find the needed > > strings by that number. > > > > 2. Add a newline between the listed problematic lines. > > At present it's tough to see what's what because everything > > is mushed together. > > I see these are now added and it indeed makes the output clearer. > > I currently have a lot of stuff to translate, maybe the script could > also show which of the incorrect strings are fuzzy, e.g. like this: > msg nr: 1487, lineno: 7704 (fuzzy) > msgid "Summary" > msgstr "%s; yhteenveto" > ? Sounds like a good idea. Checked-in. Perhaps I should add commandline options, one of which can suppress fuzzies alltogether. Sometimes you're not interested in fuzzies in the detailed report. - Kees |