This repository collects various scraps of code related to the Rare Book School class L-100 (Digital Approaches to Bibliography and Book History) in particular, and to assorted points of biblionerdery, more generally.
As will probably be clear from a quick glance at the bits and bobs you'll see here, I'm a practitioner of what could charitably be called punk rock coding: I don't always know what I'm doing, but try to make a noise anyhow. (If you actually do know what you're doing, you may well see things here that will leave you shaking your head and whistling through your teeth.)
If you see something you think you can use, please feel free. If you have questions, get in touch and I'll try to answer them, but can't guarantee that I'll know the answers myself.
This is a bit of Python meant to parse pagination statements from English Short Title Catalogue records in order to arrive at a count of the pages in a book. I wrote it for use in Open Refine. If you have a column of pagination statements in Open Refine, you could select "Edit column > Add column based on this column" and paste in the code.
The script does several things:
- It transforms all Roman numerals into Arabic numbers (its Roman numeral validation isn't as robust as some recipes you'll find online, like Mark Pilgrim's from "Dive into Python", for instance, but it seems to work), the better to do math with.
- It deals with page ranges (recognizing that "11-20" is actually 10 pages, rather than 9).
- It deals with corrections supplied for mis-numbered pages (as in "255[i.e.355]").
- It deducts numbers of plates from the page count, to leave just the letterpress pages.
- It deals with a couple of other oddities, like ignoring the numbers given for parts of books ("pts.") in favor of just page numbers, and setting the page count to 0 for any pagination statement that doesn't have any numbers--not that the book doesn't have any pages, but the pagination statement just isn't giving them to us in a usable form.
This is a work in progress--I expect there are cases I haven't thought of, so it would be prudent to scrutinize its results.
Out of the box, Zotero will ingest records from the English Short Title Catalogue quite happily, using the translator for Aleph library catalogues in conjunction with the MARC translator. The default behavior isn't especially well-suited for working with ESTC records, however. These translators make a few tweaks to improve things.
There are two javascript files here, which both need to be added to Zotero's collection of translators. English Short Title Catalogue.js notes when you're at the ESTC and directs Zotero to use the second file, MARC-2.js in place of the standard MARC.js translator. MARC-2 .jsis just a slightly altered version of the standard MARC.js translator that:
- Grabs the ESTC citation number and places it in the "Extra" field.
- Grabs the entire, messy pagination statement (rather than just the first thing that looks like a number).
- Inclues citation notes (the MARC 510 field) in Zotero notes
- Separates multiple notes with a pipe character (
' | ') for easier parsing later.
Like the ESTC translators, this translator makes a few tweaks to the existing PICA translator to make the results that Zotero imports more friendly for book historical work:
- The STCN fingerprint goes in the "Extra" field.
- The imprint information gets split into city and publisher fields.
- The format and collation statement gets put into the number of pages field.
- Typographical information gets saved as a note.
These translators lean very, very heavily on the work of the original developers, Simon Kornblith, Sylvain Machefert, Michael Berkowitz, Ming Yeung Cheung, and Sebastian Karcher. Duplicating the work of the MARC.js module for the ESTC translator strikes me as a particularly inelegant kludge, but I take some comfort in seeing that that seems to be more or less the same approach that was used to adapt the Aleph translator to deal with quirks of one library's OPAC using the MAB2.js translator.
If you're using Mac OS X, these javascript files need to go in the folder located at /Users/<your-username>/Library/Application Support/Zotero/Profiles/<your-profile>.default/zotero/translators. (You can navigate to this folder in the Finder by holding down the option key while selecting the Go menu, which will reveal the ordinarily-hidden Library folder. If you are using Windows, you'll first need to set Windows Explorer to view hidden files. If you are using Zotero Firefox, add the files to C://Users/<your-username>/AppData/Roaming/Mozilla/Profiles/<your-profile>.default/zotero/translators. If you are using Zotero Standalone, add them to C://Users/<your-username>/AppData/Roaming/Zotero/Profiles/<your-profile>.default/zotero/translators.)
I've tested these translators on Mac OS X only. They work work with Zotero Firefox and with Zotero Standalone when using Safari and Chrome. For quite some time I thought there was something keeping the translator from working with Safari, but the problem went away when I disabled all other Safari extensions and then re-enabled them. (If there was interference between two extensions, it seems to have resolved itself after a disable/enable cycle.)