[pmwiki-users] Bulk import and categories

[Michel]

Le Fri, 21 Sep 2007 05:34:51 -0500, Patrick R. Michaud a écrit :

> On Thu, Sep 20, 2007 at 07:16:52PM +0200, Michel wrote:
>> Hello,
>> 
>> I wish to import in PmWiki over than 3,500 text files created by another
>> tool.
> 
> Then you probably want to use the ImportText recipe:
> 
>     http://www.pmwiki.org/wiki/Cookbook/ImportText
> 
> This recipe makes it easy to import files, while automatically
> keeping the .pageindex and targets= lines up-to-date.  It
> also handles any problems arising from hitting PHP's
> "maximum execution time" limits.

I had a quick look at this page before, but knowing the page format and
due to the fact that I had to create a script to convert existing file, I
choosed the do-it-myself way, just for the fun (common mistake among
programmers). I came back to ImportText. I agree that this is the way to
go. Thanks for the job.

>> I have set up a script translating the pages to the Pmwiki format
>> according to the specs :
>> http://www.pmwiki.org/wiki/PmWiki/PageFileFormat
> 
> We should probably add a link to the PmWiki.PageFileFormat page that
> points people to the Cookbook:ImportText option.  It's probably also a
> good idea to put this as a FAQ in the PmWiki.FAQ page.  (I'd do it now
> but I'm on a job site this week and can't easily get to it until the
> weekend.)

Perhaps you can add some warning explaining why ImportText is preferable.
Here are the arguments from that I have just learned. Please use/correct
as needed.

- direct import create incoherence between pages and internal indexes, and
does not manage categories or trail ;

- if you really want to direct import files without Category/trail, you
need to delete .pageindex. The file will be rebuilt when needed.
Caution : depending of the number of pages, rebuilding can be aborted due to PHP
time-out.

- if you really want to direct import files with Category/trail, you need
to insert a 'targets=' field to you pages before importing.

You can avoid all these problems in using Importtext :-)

Thanks a lot

Michel