This coding week has been really hard. My laptop simply died when I was coding last Saturday night. It turned on unexpectedly last Monday so I could easily backup my Friday and Saturday work which included a Bibtex parser. But the problem persists: it sometimes works and sometimes don't. It is obvious that I definetily cannot rely on it so I brought it to a technical center and had to borrow my father's laptop which I cannot use every moment as he needs it too. I used my Monday to install the things I need to continue the development but two days wasted is really painful for me. I am a bit dissapointed because now I am behind schedule.

But life continues and JResearch is going on. Importing functionalities are complete including the interface to upload files in Bibtex, RIS and MODS formats. As Neri exposed in her last blog, I made a change at code level related to the native format supported by JResearch. I stated that would use MODS, as it is an intermediate XML format supported by Bibutils package. However when testing conversions from Bibtex to MODS, I realized a slight loss of information, specifically in the publication type. Booklets records were converted to Books and there is no difference between PhdThesis and Masterthesis. The problem: my database was designed thinking in Bibtex records types, so I decided that everything would be converted to Bibtex first and then parsed. The results were satisfactory even though parsing MODS would have been easier (it is XML). As JResearch relies on Bibutils, I have provided the binaries for Windows, MAC and Unix (i386 platforms). I think that compiling them in other platforms will not be a problem.

JResearch Control panel is also ready, thanks to Neri's contribution in the part of art. I think it looks elegant.

Component configuration form is also ready thanks to the amazing com_config. Getting a functional and elegant configuration page is as easy as writing a XML file. I am amazed.

Cleaning session problem is like a headache. Cited records, invoked when writing an article are kept in session, but I have not been able to find a way of cleaning the session when the user has saved the item as there is not an event onSaveArticle. Additionally, I have decided to implement persistency, so the user can get those cited records whenever editing the item. I tried implementing listeners for several events (via plugins) with no success. A better explanation of the problem can be found in (http://forum.joomla.org/viewtopic.php?f=304&t=306402). The idea is to store them in a table that contains the id of the item, the type of item (article for com_content records, publications || projects || theses for com_jresearch) and the publication's citekey. I can manage the event onSaveArticle when saving JResearch entities, but that is not possible for com_content. Neri recommend me to check user and registry store capabilities, but now the problem is to match the cited information with the id of the item. What happens when the the item is new (it does not have an id)?. Important to mention that the user can cite publications via editors-xtd and native TinyMCE plugins. Please help need here!!!!

APA citation style is implemented. One function is missing and needs to be tested. MLA is understood and ready to be implemented.

I am behind schedule, so I need to work hard this weekend. My motivations:

- MLA and APA citation styles complete and tested for Monday 21st.
- Issues associated to importers functionality testing
- Administrative list of publications view, for Wednesday 23rd
- Exporters hierachy and views for editing/creating articles and books for Monday 28th. Note that publications have a set of common fields. A lot of code will be reused.
- Projects, theses, publications and research areas backend ready by August 4th.

Again, help and feedback are always welcome.