Re: [PEPr] Comment on File Formats::File_CSV_Get
| From: | till | Date: | Mon, 17 Nov 2008 11:32:46 +0000 |
| Subject: | Re: [PEPr] Comment on File Formats::File_CSV_Get | ||
| References: | 1 2 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-51102@lists.php.net to get a copy of this message | ||
On Mon, Nov 17, 2008 at 12:23 PM, Kazuyoshi Tlacaelel
<kazu.dev@gmail.com> wrote:
>
> Kazuyoshi Tlacaelel (http://pear.php.net/user/ktlacaelel) has commented on the proposal for
> File Formats::File_CSV_Get.
>
> Comment:
>
> Till, Thanks for you comments.
> You have a great attitude, cheers!
>
>
>> I checked out your code and looked at it very briefly, ran tests -
> great
>> documentation, great examples.
>
> Thanks, there is more to do, but right now I though of making the
> proposal
> and make some changes to the code before proceeding with documentation ;
> )
>
>> There are small things I'd like you to change, for example, using
>> protected instead of private so people can extend/overload your class
>
> I agree with you, good observation!
>
> In the beginning I though of making them private so people don't start
> using something that may change.
There's no need to do this.
Assuming you start with 0.1.0-alpha, or even 0.0.1-alpha you have
plenty of opportunities to change your API until you hit a
1.0.0-stable. Just document the changes when you do a release on PEAR.
It's not like BC breaks are encouraged (I'd say), but they are allowed
until you go major version (aka 1.0.0).
> I'll keep the patch methods private, and change some methods visibility!
>
>> If you keep as-is, think about file_exists() vs. is_readable(). I
> *think*
>> on some systems there can be a difference, which can lead to a nasty
> error.
>> ;-)
>
> Yeap, sounds good, I'll apply this changes too.
> I also like the idea of the url, I will start thinking on that as well.
> yes! the file extension is supposed to go!
>
> I added some you proposals to the issues on google-code
> and will come back to this post to see if I missed something.
> http://code.google.com/p/php-csv-parser/issues/list
Sweet. ;-)
> About the svn file structure..
> What should I change?
> I had my own doubts about that, about
> - "tests/data" going to "data" instead
> - and misc.
File_CSV_Get
File/
File/CSV/
File/CSV/Get.php
tests/
examples/
package.xml
Check this as an example:
http://cvs.php.net/viewvc.cgi/pear/Services_Blogging/
> I normally just check out trunk and rename it to "CSV"
> under a "File" directory that is not under subversion.
>
> then under "CSV" run
>
> $ sudo ./build
>
> This unistalls the package.
> runs "phpcs", "phpdoc", "phpunit" adds phpdocumentor compiled
> docs
> to package.xml and builds the package, terminating by re-installing it,
> and logging the results.
>
> Is there any task that runs "all pear-projects test-code" ?
> should I change my test structure to match that?
Yeah...
pear run-tests -u tests/
-u if you are using phpUnit.
To validate your package.xml:
pear package-validate
To create a package from your current package.xml
pear package
Regards,
Till