[Okular-devel] Re: MD5 document identification

[Miguel Vazquez]

  On 10/13/10 12:28 AM, Albert Astals Cid wrote:
> A Divendres, 8 d'octubre de 2010, Miguel Vazquez va escriure:
>>    Well, that would mean saving the .okular file into a particular
>> directory on my filesystem and then open that version next time.
>> However, I maintain the pdf repository online and I'd like to be able to
>> open the pdf by clicking the links on the online version, which would
>> download a pdf new file into /tmp/ and would not be recognized by okular
>> as having annotations.
>>
>> My fix  is to save each pdf into a particular directory first time I
>> open it, and annotations will get associated with the path of that
>> version. Next time I open it I will detect I have the same file in the
>> repository, by md5, and open that version instead (roughly). The problem
>> is that I need to save an extra version of the file in that directory. I
>> guess I could work around actually saving each file with a little work,
>> but it will all be simpler if okular would make the md5 check itself.
> Well, in the past we got complains when we did not use the path for
> annotations since people said it was a different file, and now that's why we
> use the path as identifier, adding the feature you request would annoy also
> these group of people, i guess one can't make everybody happy.
>
> Albert
>

Fair enough.

Cheers

Miguel
>> Anyway, its just a suggestion, it may very well not make sense to
>> include that functionality.
>>
>> Thank you very much for developing this tool, I found it extremely useful.
>>
>> Cheers.
>>
>> Miguel
>>
>> On 10/07/2010 08:23 PM, Albert Astals Cid wrote:
>>> A Dijous, 7 d'octubre de 2010, Miguel Vazquez va escriure:
>>>>     Hi,
>>> Hi!
>>>
>>>> I built a system to centralize my annotations for a document collection
>>>> I have of scientific articles. I like to download them from a central
>>>> document repository online and have them use any annotations I had from
>>>> a previous read. I used the attached script that uses md5 checksums to
>>>> see if the articles is already in my repo and uses that instead of the
>>>> downloaded one (the script might not work straight out of the box
>>>> though).
>>>>
>>>> Any way, maintaining my own pdf repo is a bit of a waste, I was
>>>> wondering what you think of associating annotations with md5 checksums
>>>> or something like that instead of file paths. For bookmarks
>>>> unfortunately this does not work, because you need the path to open the
>>>> file.
>>>>
>>>> Any thoughts?
>>> Doesn't saving as document archive (.okular) fix this problem?
>>>
>>> Albert
>>>
>>>> Miguel
>>> _______________________________________________
>>> Okular-devel mailing list
>>> Okular-devel at kde.org
>>> https://mail.kde.org/mailman/listinfo/okular-devel
>> _______________________________________________
>> Okular-devel mailing list
>> Okular-devel at kde.org
>> https://mail.kde.org/mailman/listinfo/okular-devel
> _______________________________________________
> Okular-devel mailing list
> Okular-devel at kde.org
> https://mail.kde.org/mailman/listinfo/okular-devel