Issues

ZF-10177: Misspelled and duplicated validation messages

Description

Unhappy with the default translations available for the various validators of the Framework, I decided to give it a go myself and I found a few inconsistencies while listing all the messages.

  • There is an unconstistent use of final punctuation. Some messages have a final punctation, some don't, it doesn't really matter, but it feels a bit cluttered.
  • There is an unconsistent use of quotes around values, for exemple you have : {quote}{{'%value%' is less than {color:blue} %min%{color} characters long}} {{'%value%' is not greater than {color:blue}'%min%'{color}}}{quote}

  • But more importantly, there is subtle duplicated messages everywhere. There may be more than the few examples I'm going to cite here, but I can't possibly find every one of them. I guess an IDE could be a great help looking for these. {quote}{{File '%value%' could not be found}} {{File '%value%' does not exist}} {{File '%value%' was not found}}{quote} {quote}{{Invalid type given, the value should be a string}} {{Invalid type given, value should be a string}} {{Invalid type given, value should be string}}{quote} {quote}{{Invalid type given, the value should be a integer, a float or a numeric string}} {{Invalid type given, value should be float, string, or integer}} {{Invalid type given, value should be string, integer or float}}{quote} {quote}{{Invalid type given, value should be string or integer}} {{Invalid type given. The value should be a string or a integer}}{quote} {quote}{{'%value%' must contain only digits}} {{'%value%' contains characters which are not digits; but only digits are allowed}}{quote} {quote}{{The service returned a failure while validating '%value%'}} {{Validation of '%value%' has been failed by the service}}{quote} {quote}{{The mimetype of file '%value%' could not be detected}} {{The mimetype of file '%value%' could not been detected}}{quote}

While I understand this does not stand as an important failure, it does look unprofessional and is a bit embarassing for us since we have to send duplicates to the translation company we work with, and obviously, they bill their services by the line (and we can't remove them manually, it's part of a much larger automated process we don't really want to mess with).

Comments

So you have a massive problem because 20 or 40 lines from the 150 provided lines are duplicated. This makes about 30 to 50,-- $ for a translation company.

I expect that you have much more translations than those few lines. And those lines ARE already translated within 12 languages. I see no "translation" problem at all especially since you dont need to re-translate already translated content.

Additionally I checked the last two just to see if there are really inconsistences within the translation... but I can't find the second one.

I expect that you are using an outdated release as you did not note which release you are using. This means for me that this issue is partitially already fixed which makes it not a bug but only an improvement in my eyes. Especially as everyone can change these translations to whatever he wants to have outputted.

Removing affected version as some noted and checked strings are already fixed within 1.10.6

Checking from above to bottom:

  • There is only ONE sentence with a final punctation. Not several or more... only one single message
  • There are only 2 sentences with a quoting which is discussable.

But your example is incorrect:

It is wished to get:


And it's also wished to get: 

Both given examples are wished and consistent.

You sound so condescending and arrogant, that's really a pain. Last time I reported an issue, you behave likewise and in the end, you admitted you were wrong and fixed it anyway. So please, stop being such a smart-ass and try to actually take a few minutes to understand people's problems before assuming people are morons. For comprehension's sake, I'm gonna reply to you the same way you replied to me, condescending and arrogant.

As I already stated in my original message, I don't use the default translations because... well... it's nothing like understandable for real non-IT-capable customers. Don't tell me final users should have to understand concepts like Zend_Date, array or Luhn algorithms. They don't care about this. Which is why I'm rewriting new translations for the default messages (and also because we don't want to use multiple formats for the translation files and we're not actually using PHP files). Maybe I'm doing it wrong, maybe I'm doing it right, I'm just following the documentation:

  • http://framework.zend.com/manual/en/… Now if you don't want people to do this, stop writing they can do it in the documentation. {quote}Note: Supported languages This feature is very young, so the amount of supported languages may not be complete. New languages will be added with each release. Additionally feel free to use the existing resource files to make your own translations. You could also use these resource files to rewrite existing translations. So you are not in need to create these files manually yourself.{quote}

Then you say: {quote}Additionally I checked the last two just to see if there are really inconsistences within the translation... but I can't find the second one.{quote} It took me one grep: {quote}root@/home/nicolas/projects/zend# grep -R "could not been" . ./library/Zend/Validate/File/IsCompressed.php: self::NOT_DETECTED => "The mimetype of file '%value%' could not been detected",{quote}

Then you say: {quote}I expect that you are using an outdated release as you did not note which release you are using{quote} I did forgot to specify which version I was using, however, I also did specify the issue was still present for the next minor version since I'm using the trunk and these issues come from the trunk. Now of course, you removed that part from my bug report, so I guess that was irrelevant.

Then you say: {quote}Removing affected version as some noted and checked strings are already fixed within 1.10.6 {quote} And then the documentation says: {quote}[...] as of 1.6.0, you may provide translation strings using the actual error messages as message identifiers. This is the preferred use case for 1.6.0 and up, as we will be deprecating translation of message keys in future releases.{quote} And then I'm like, WTF? If we start using actual error messages as message identifiers, I hope you're not going to change them in each minor version, this is quite hard to keep up and that's why I started to list all the messages that seemed misspelled or duplicated... Now I understand you don't care because obviously, this is just a massive problem because 20 or 40 lines from the 150 provided lines are duplicated. That's OK, I'm not gonna bother you anymore, just do whatever you want with your framework but then don't protest if people stop reporting issues.

{quote}You sound so condescending and arrogant...{quote} {quote}stop being such a smart-ass{quote} {quote}I'm gonna reply to you the same way you replied to me, condescending and arrogant{quote}

I am no native english speaking person and my english may not be very good. But there is no need to curse at people.

When you have a problem, to communicate with non english people, in a way that you must use dirty words when you are getting a reply then I would keep in mind that this is not very constructive. Almost all people here, including me, are spending their sparetime to work on this framework.

Such a reply is neither positive, nor constructive, nor useful. There will be no further response. Work in done independently from this issue.

I am closing this issue because of the vulgar response.

Reopening because (1) there is no vulgar response and (2) we have no policy of closing issues containing vulgar wordings. Thomas, I think the reporter really didn't want to talk down to you, and in fact didn't do so.

Fixed

Thank you for fixing this, even if we had a hard time understanding each others.

Setting fix version to 'next minor release' rather than 'next mini release'.