A CProject is set up by a local user or a program run by them and will contain CTrees . Generally this maps onto operations such as:
- downloading one or more papers (+ supplemental) with
getpapersorquickscrape. - processing a collection of papers already on a user's disk.
A CProject will often have a purpose such as:
- a daily trawl through the literature (e.g.
daily_20150102) - a retrospective crawl of a publisher (
bmc2012-4) - a query (
brontosaurus_2013)
- `CProject's are based on standard filesystem directories.
CProjects can be set up anywhere on a users disk.- ISSUE. SHOULD they have reserved syntax? e.g.
c_foo - ISSUE.
CProjectfilenames should be <= 20 characters and contain only `[A-Za-z][A-Za-z0-9_]. This is to prevent spaces, OS-dependent chars, and support processability with software. Also long filenames may cumulatively be problematic for ZIP. CProjectMUST have amanifest(the files they contain and when they were created/modified)CProjectMUST have an audit trail (including provenance for legal purposes - when, from where and by what were they created)CProjectMAY contain 0 or moreCTrees- ISSUE
CProjectMAY / MUST_NOT contain files other than the above. - ISSUE
CProjectMUST NOT containCProject.
- from a
getpaperssearch - from a
quickscrapeon a single URL - from a
quickscrapewith a list of URLs or a generator of URLs - from a list of files on disk (e.g.
*.pdf). Usingnorma. Files must all be of the same type (*.pdf) OR *.html or *.txt - as an empty `CProject.