Published data
Published data (DOI)
This software's choice — A decision made by this software rather than by the field's sources; the explanation says why it was made.
The DOI source opens a Zenodo or figshare record by its DOI and lists the files it publishes.
Paste a DOI - 10.5281/zenodo.1234567, or the same thing as a doi.org link - and Fit asks the DOI system where it resolves to, then asks that repository which files the record holds. Next goes straight to that list: choose a file and it is previewed and imported like any other.
Words rather than a DOI search Zenodo and figshare for datasets whose description matches them - datasets only, not papers or software, and only those holding a file Fit can read. Each row says which repository it is in, when it was published and how many of its files Fit can read. Choose one and press Next to see its files; Back returns to the records.
A page holds up to 25 records from Zenodo - the most it gives without an account - and ten from figshare, whose records are each asked for their files before they are listed. A row "More from Zenodo..." or "More from figshare..." at the end opens that repository's next page.
When one repository cannot be reached, the other's records are still listed, with a row saying which could not be searched; opening that row says why. Only both failing stops the search.
Only what Fit can read is listed. A record's papers, pictures and anything else no reader here understands are left out, so the list is the data you can fit; a record holding nothing of the kind says how many files it did hold, rather than showing an empty list. Which files count follows the readers this build has: a module that adds a format adds its files here too.
An archive - .zip, .gz, .tar, .tar.gz or .tgz - is listed as a folder. Opening it downloads it once and lists the files inside that Fit can read, an archive inside it as another folder; choosing one takes that file alone, named as the archive names it. Nothing is unpacked onto the disk.
Limitations
- Only Zenodo and figshare publish a file list Fit can read. A DOI that resolves anywhere else is refused, naming where it went, so that the file can be fetched with the Web address source instead.
- An archive is unpacked in memory, and a file in it larger than 256 MB is refused rather than unpacked. A .zip that is encrypted, or split over several files, cannot be opened.
- A record may publish its data in a form no reader here understands - an instrument's own binary format, say. Such files are not listed; open the record in a browser to see everything it holds.
See also: Data catalog · Data sources
Data catalog
This software's choice — A decision made by this software rather than by the field's sources; the explanation says why it was made.
The Data catalog browses the newest datasets of Zenodo and figshare by the kind of data Fit can read, like folders, with nothing to type.
Choosing it lists one row for each kind of file this build reads - a diffractometer's scans, spectra, tables of numbers, workbooks and the rest, each with its extensions - and a row for archives. Opening a row lists the newest datasets on both repositories holding such a file; opening a dataset lists the files Fit can read, and an archive among them opens like a folder. A "More from..." row at the end opens the next page. Back goes up a level.
Nothing here is a list kept by Fit: the rows are the readers this build has, so a module that adds a format adds a row, and everything under them is asked of the repositories when it is opened.
Limitations
- Plain .txt files are no row of their own, although they are read: nearly every dataset holds a README.txt, and the row would list them all. A .txt is still listed inside a dataset, and a word search with Published data (DOI) still finds datasets by them.
- A dataset is listed for the name of a file it holds; whether that file is a measurement is seen in the preview.
- How many datasets a row holds is not shown: counting would ask both repositories once for every row before the list could be drawn.
See also: Published data (DOI) · Data sources