Hi Egon, Pantelis, All,

On 15 July 2010 20:41, Egon Willighagen <egon.willighagen at gmail.com> wrote:
> On Thu, Jul 15, 2010 at 6:34 PM, Vedrin Jeliazkov
> <vedrin.jeliazkov at gmail.com> wrote:
>> Introducing printer-friendly & fancy images has some pros and cons:
>>
>> + could improve the overall design, look and feel of the (printed) report;
>> - opens the door for data inconsistencies (e.g. a fancy image which
>> does not correspond to the chemical representation of the compound --
>> SMILEs, InChI or whatever -- would be rather useless and could result
>> in further errors, made by the person, using this report);
>
> This could be addressed by putting the InChI and/or CML in the PNG image itself:
>
> http://chem-bla-ics.blogspot.com/2007/08/molecular-connectivity-tables-in-images.html
> http://neksa.blogspot.com/2007/08/strigi-now-extracts-chemical.html
>
> This basically hard-codes what the PNG is about.

This would be a nice solution, indeed. I have a few more questions, though:

1) Do you know by any chance some library (preferably, but not
necessarily written in Java) providing methods for encoding and
retrieving chemical structure information in images?

2) Are you aware of any standardization work on how such information
should be encoded in images? (this might be important if we're looking
for interoperability with 3rd party software)?

3)  Any further progress on including InChIs?

Kind regards,
Vedrin