4. Getting the PDFs

mosaic downloads what you can read legally for free, and what your own access opens (chapter 5). For papers whose source gave no PDF link it asks Unpaywall, which wants an e-mail address with every request. Set yours once:

mosaic config --unpaywall-email you@example.org

Without it, mosaic skips Unpaywall and says so when a download fails.

While searching

$ mosaic search "hall thruster particle-in-cell" --source arxiv --max 3 --download

 #     Title                                  Authors         Year   DOI                    Source         OA     PDF
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
 1     Plasma Diagnostic and Performance of   Jose Leonardo   2004   10.48550/arXiv.physi   arXiv          yes    ✓
       a Permanent Magnet Hall Thruster       Ferreira et            cs/0410170
                                              al.
 2     Reduced-order particle-in-cell         Maryam Reza     2023   10.1088/1361-6595/ac   arXiv          yes    ✓
       simulations of a high-power            et al.                 dea3
       magnetically shielde…
 3     Effects of the neutral dynamics        Farbod Faraji   2023   10.1063/5.0143292      arXiv          yes    ✓
       model on the particle-in-cell          et al.
       simulations of a …
3 result(s)
  ✓ 2004_arXiv_Jose_Plasma_Diagnostic_and_Performance_of_a_Permanent_Magnet_Hall.pdf
  ✓ 2023_arXiv_Maryam_Reduced-order_particle-in-cell_simulations_of_a_high-power_m.pdf
  ✓ 2023_arXiv_Farbod_Effects_of_the_neutral_dynamics_model_on_the_particle-in-cel.pdf


Done: 3 downloaded, 0 failed, 0 skipped (no PDF)

--download (-d) fetches the PDF of every result shown, into ~/mosaic-papers/. A paper downloaded before is not fetched again: the cache remembers where it is. --download-dir DIR sends one run elsewhere; mosaic config --download-dir DIR changes the default.

One paper by DOI

$ mosaic get 10.3389/fphy.2019.00140
Numerical Study of Electron Cyclotron Drift Instability: Application to Hall Thruster (2019)
Saved: ~/mosaic-papers/2019_DOAJ_Zahra_Numerical_Study_of_Electron_Cyclotron_Drift_Instability_Appl.pdf

mosaic get takes a DOI, bare, with a doi: prefix or as a https://doi.org/ link; an arXiv paper has the DOI 10.48550/arXiv.<id>. mosaic looks for a PDF in this order and stops at the first file that really is a PDF:

1. the PDF link in the cache, if a search found one;
2. every open-access copy Unpaywall knows, best first: the publisher's, a repository such as arXiv, PubMed Central, HAL or a university archive;
3. a browser session you saved with mosaic auth login (chapter 5).

An open-access paper can still fail:

$ mosaic get 10.3390/aerospace10010044
PIC-DSMC Simulation of a Hall Thruster Plume with Charge Exchange Effects Using pdFOAM (2023)
Could not find a downloadable PDF for this DOI.
Have access? Retry with --handoff (your own browser; works for Elsevier) or --interactive (a browser window mosaic
drives).

MDPI answers every scripted download with HTTP 403, an anti-bot wall, and no repository holds a copy of this paper. mosaic does not try to get past such walls; the hint names the two ways that use your own browser.

The library

$ ls ~/mosaic-papers
2004_arXiv_Jose_Plasma_Diagnostic_and_Performance_of_a_Permanent_Magnet_Hall.pdf
2019_DOAJ_Zahra_Numerical_Study_of_Electron_Cyclotron_Drift_Instability_Appl.pdf
2023_arXiv_Farbod_Effects_of_the_neutral_dynamics_model_on_the_particle-in-cel.pdf
2023_arXiv_Maryam_Reduced-order_particle-in-cell_simulations_of_a_high-power_m.pdf

Files are named {year}_{source}_{author}_{title}.pdf. {author} is the first word of the first author's name as the source wrote it: a given name from arXiv ("Maryam Reza"), a family name from Crossref ("Reza, Maryam"). Titles are cut to 60 characters. Change the pattern with mosaic config --filename-pattern; the placeholders are {year}, {source}, {author}, {title}, {journal} and {doi}.

A list of DOIs

Your reference manager exports BibTeX; mosaic reads the doi fields of a .bib file, or the doi column of a .csv:

$ mosaic get --from refs.bib
Found 2 DOI(s) in refs.bib
  ✗ 10.1063/1.4972269
  ✓ 2019_Crossref_Taccogna_Latest_progress_in_Hall_thrusters_plasma_modelling.pdf


Have access? Retry with --handoff (your own browser; works for Elsevier) or --interactive (a browser window mosaic
drives).

Done: 1 downloaded, 1 failed

For each DOI mosaic takes the title, authors and year from the cache or from Crossref, so the files get the same names as after a search. The first paper has no open copy anywhere; it waits for chapter 5.

Checking the library

$ mosaic cache verify
5 file(s) OK, 0 missing, 0 not a PDF

mosaic cache verify checks that every recorded download still exists and starts with the PDF signature. A web page saved as .pdf (a login or bot-check page, written by versions before 1.6) counts as "not a PDF". mosaic cache clean forgets those records, and the missing ones, so the next get tries again; it deletes no file.

What you learned

--download, mosaic get with a DOI or a .bib/.csv file, where mosaic looks for a PDF, file names, cache verify and cache clean. Reference: get, guide: Downloads.

Next: 5. Paywalled papers.