Comments on: Zotero, XML, Python, and SP28 http://www.investigations.4-lom.com/2007/12/20/zotero-xml-python-and-sp28/ Failing better at understanding the past Thu, 20 Nov 2008 15:43:22 +0000 http://wordpress.org/?v=2.6.2 By: Gavin Robinson http://www.investigations.4-lom.com/2007/12/20/zotero-xml-python-and-sp28/#comment-11628 Gavin Robinson Sun, 23 Dec 2007 20:42:05 +0000 http://www.investigations.4-lom.com/2007/12/20/zotero-xml-python-and-sp28/#comment-11628 I know <a href="http://digitalhistoryhacks.blogspot.com/" rel="nofollow">Bill Turkel</a> has mentioned going directly into the Zotero SQLite database, but I haven't tried it myself. In this case I'm putting in custom XML as notes anyway, and I find XML very easy to work with, so exporting the records as XML seemed like the way to go. I'm now increasingly thinking that even with custom fields it won't be practical to use Zotero to store all the data I want. At least some documents are just too big and complicated and will need to be represented in a custom database or XML. For example, the account book I photographed at the PRO last month has around 3,000 entries! I know Bill Turkel has mentioned going directly into the Zotero SQLite database, but I haven’t tried it myself. In this case I’m putting in custom XML as notes anyway, and I find XML very easy to work with, so exporting the records as XML seemed like the way to go.

I’m now increasingly thinking that even with custom fields it won’t be practical to use Zotero to store all the data I want. At least some documents are just too big and complicated and will need to be represented in a custom database or XML. For example, the account book I photographed at the PRO last month has around 3,000 entries!

]]> By: Mike Cosgrave http://www.investigations.4-lom.com/2007/12/20/zotero-xml-python-and-sp28/#comment-11625 Mike Cosgrave Sun, 23 Dec 2007 19:49:40 +0000 http://www.investigations.4-lom.com/2007/12/20/zotero-xml-python-and-sp28/#comment-11625 This is very useful stuff! However, since Zotero uses sqlite as its database, surely you can use any sqlite front end to pull out the records you want directly into a separate database in order to perform analysis? For some research, this could be handier than exporting and re-importing the data. I wonder if OpenOffice Base can connect to SQLite? Then you could just populate a Calc sheet with the data, and work out the pounds-shillings-pence cost of supplies, regardless of whether they were delivered in tuns, kilderkins, pecks barrels or bushels! Maybe I'll get to play with these over the holidays. This is very useful stuff!

However, since Zotero uses sqlite as its database, surely you can use any sqlite front end to pull out the records you want directly into a separate database in order to perform analysis? For some research, this could be handier than exporting and re-importing the data.

I wonder if OpenOffice Base can connect to SQLite? Then you could just populate a Calc sheet with the data, and work out the pounds-shillings-pence cost of supplies, regardless of whether they were delivered in tuns, kilderkins, pecks barrels or bushels! Maybe I’ll get to play with these over the holidays.

]]> By: Gavin Robinson http://www.investigations.4-lom.com/2007/12/20/zotero-xml-python-and-sp28/#comment-11407 Gavin Robinson Fri, 21 Dec 2007 10:59:37 +0000 http://www.investigations.4-lom.com/2007/12/20/zotero-xml-python-and-sp28/#comment-11407 Actually that function to get the text from an element doesn't work properly if there isn't any text, and it's probably unnecessary anyway. At first I was just using firstChild to extract the text, but then I thought "what if there happens to be more than one child node?". Well, with the custom XML I'm using the elements I'm interested in won't have any children other than the text they contain, and if they do it means I've done something wrong. Maybe in future I'll have to think of something more sophisticated but for now it seems like I'm trying to solve a problem that I don't really have. Actually that function to get the text from an element doesn’t work properly if there isn’t any text, and it’s probably unnecessary anyway. At first I was just using firstChild to extract the text, but then I thought “what if there happens to be more than one child node?”. Well, with the custom XML I’m using the elements I’m interested in won’t have any children other than the text they contain, and if they do it means I’ve done something wrong. Maybe in future I’ll have to think of something more sophisticated but for now it seems like I’m trying to solve a problem that I don’t really have.

]]>