Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Save/Restore tabs #6

Open
0xdevalias opened this issue Aug 28, 2018 · 3 comments
Open

Save/Restore tabs #6

0xdevalias opened this issue Aug 28, 2018 · 3 comments
Labels

Comments

@0xdevalias
Copy link
Owner

0xdevalias commented Aug 28, 2018

As I tab junkie I tend to end up with a lot of tabs related to certain 'contexts', and this extension was born out of helping me to 'slice off' those contexts, so that I could then save them for future using something like Sesh (Chrome Store, GitHub)

I'm not sure if this would necessarily make sense to include in this extension.. sort of torn between 'keep it simple' and 'integrate the features I need so I don't require multiple extensions'. Described here for future pondering.

One potential options is to have some features that 'play nice' with Sesh's features, so that by using both you get a 'synergistic effect'. This concept could potentially be extended to other extensions as well, which would reduce the need to 'reimplement features', and help support others work (eg. This features requires you to also have X extension installed: (link to chrome store))

API's

@0xdevalias
Copy link
Owner Author

0xdevalias commented Jul 2, 2022

Note that Sesh uses Chrome's 'Other Bookmarks' folder as the default root/parent to save its sessions in. According to the extension code, there is no simple way to get the ID of this folder except by looking it up by name, which will differ depending on the browser's language settings. The extension therefore includes a bunch of unicode escaped translations by default:

https://github.com/yujiberra/sesh/blob/7c0ab6a9ff07df40fc3af2a64d0c97593f9b4530/sesh.js#L127-L185


A little google-fu lead to me finding the 'Other Bookmarks' entry in the Chromium translation files:

<translation id="883848425547221593">Other Bookmarks</translation>

Which can also be found in the GitHub mirror of the Chromium source code:

We can see all of the translated resource files here, in the generated_resources_*.xtb files:

Currently, by rough manual count, there seem to be ~81 different translation files (each of which will presumably have their own variation of how 'Other Bookmarks' is shown in the UI.. though there may be duplicates)

Googling for the translation id (883848425547221593) pointed to a few different translation files seeming to hold the 'Other Bookmarks' string in other Chrome related repos:

Googling for this new translation id (2354001756790975382) turned up a few more interesting looking results:

Which still seem to be present in the latest source:

And can be found on the GitHub mirror of the repo in the components_strings_*.xtb files:


We can git clone with a few arguments to only retrieve the files we want (rather than the whole repo):

Doing a --depth=1 clone of the chromium source results in ~1.33GB of largely unneeded files being downloaded, that results in ~5.8GB of files on disk:

⇒  git clone --depth 1 https://github.com/chromium/chromium
Cloning into 'chromium'...
remote: Enumerating objects: 403624, done.
remote: Counting objects: 100% (403624/403624), done.
remote: Compressing objects: 100% (283941/283941), done.
remote: Total 403624 (delta 110999), reused 283472 (delta 98263), pack-reused 0
Receiving objects: 100% (403624/403624), 1.33 GiB | 12.27 MiB/s, done.
Resolving deltas: 100% (110999/110999), done.
Checking connectivity: 403624, done.
Updating files: 100% (400650/400650), done.

⇒  du -sh chromium
5.8G	chromium

Using the 'partial clone' hack from above...

⇒  git clone --depth 1 --filter=blob:none --sparse https://github.com/chromium/chromium
Cloning into 'chromium'...
remote: Enumerating objects: 34624, done.
remote: Counting objects: 100% (34624/34624), done.
remote: Compressing objects: 100% (25767/25767), done.
remote: Total 34624 (delta 1785), reused 21959 (delta 1265), pack-reused 0
Receiving objects: 100% (34624/34624), 12.97 MiB | 11.89 MiB/s, done.
Resolving deltas: 100% (1785/1785), done.
remote: Enumerating objects: 27, done.
remote: Counting objects: 100% (27/27), done.
remote: Compressing objects: 100% (27/27), done.
remote: Total 27 (delta 2), reused 6 (delta 0), pack-reused 0
Receiving objects: 100% (27/27), 191.39 KiB | 7.97 MiB/s, done.
Resolving deltas: 100% (2/2), done.

⇒  du -sh chromium
 72M	chromium

⇒  cd chromium

⇒  git sparse-checkout set components/strings
remote: Enumerating objects: 299, done.
remote: Counting objects: 100% (299/299), done.
remote: Compressing objects: 100% (170/170), done.
remote: Total 299 (delta 166), reused 156 (delta 128), pack-reused 0
Receiving objects: 100% (299/299), 6.22 MiB | 6.25 MiB/s, done.
Resolving deltas: 100% (166/166), done.
Updating files: 100% (326/326), done.

⇒  du -sh ../chromium
106M	../chromium

⇒  du -sh components/strings/
 29M	components/strings/
 
 ⇒  du -shc components/strings/components_strings_*.xtb
280K	components/strings/components_strings_af.xtb
..snip..
296K	components/strings/components_strings_zu.xtb
 28M	total

greping for the translation IDs in these files, we can see that since there are 2 different IDs, we get 2 entries per file:

⇒  grep -r -E '2354001756790975382|883848425547221593' components/strings
components/strings/components_strings_fr.xtb:<translation id="2354001756790975382">Autres favoris</translation>
components/strings/components_strings_fr.xtb:<translation id="883848425547221593">Autres favoris</translation>
..snip..
components/strings/components_strings_si.xtb:<translation id="2354001756790975382">වෙනත් පිටු සලකුණු</translation>
components/strings/components_strings_si.xtb:<translation id="883848425547221593">වෙනත් පිටු සලකුණු</translation>

⇒  grep -r -E '2354001756790975382|883848425547221593' components/strings | wc -l
    162

But since both translation IDs seem to result to the same string, we can probably just pick one of them:

⇒  grep -r '883848425547221593' components/strings | wc -l
      81

⇒  grep -r '2354001756790975382' components/strings | wc -l
      81

⇒  grep -r '2354001756790975382' components/strings
components/strings/components_strings_fr.xtb:<translation id="2354001756790975382">Autres favoris</translation>
..snip..
components/strings/components_strings_si.xtb:<translation id="2354001756790975382">වෙනත් පිටු සලකුණු</translation>

We could also use xpath to extract these 'properly', but it seems way slower than grep (~40sec vs ~0.5sec):

time xpath -q -e '//translation[@id='2354001756790975382']/text()' components/strings/components_strings_*.xtb | wc -l
      81
xpath -q -e '//translation[@id='2354001756790975382']/text()'   40.75s user 0.38s system 94% cpu 43.389 total
wc -l  0.00s user 0.00s system 0% cpu 43.387 total

⇒  time grep -r '2354001756790975382' components/strings | wc -l
      81
grep --color -r '2354001756790975382' components/strings  0.47s user 0.04s system 93% cpu 0.539 total
wc -l  0.00s user 0.00s system 0% cpu 0.538 total

But using a little shell magic, we can combine the two and get pretty instant results:

⇒  TRANSLATIONS=`grep --no-filename -r '2354001756790975382' components/strings` && echo -e "<translations>\n$TRANSLATIONS\n</translations>" | xpath -q -e '//translation/text()' | wc -l
      81

⇒  TRANSLATIONS=`grep --no-filename -r '2354001756790975382' components/strings` && echo -e "<translations>\n$TRANSLATIONS\n</translations>" | xpath -q -e '//translation/text()'
Autres favoris
..snip..
වෙනත් පිටු සලකුණු

@0xdevalias
Copy link
Owner Author

0xdevalias commented Jul 6, 2022

We could also achieve the same sort of thing as #6 (comment) by using the GitHub API rather than cloning the git repo directly:

The following command uses the gh CLI program to query for the files in the components/strings/ directory of the chromium/chromium repo, filter the result JSON for files that match components_strings_*.xtb, then get the .download_url string from each JSON object in the result array:

⇒  gh api /repos/chromium/chromium/contents/components/strings --jq 'map(select(.type == "file" and (.name | test("components_strings_.+\\.xtb"; "i")))) | map(.download_url) | .[]' | wc -l
      81

⇒  gh api /repos/chromium/chromium/contents/components/strings --jq 'map(select(.type == "file" and (.name | test("components_strings_.+\\.xtb"; "i")))) | map(.download_url) | .[]' | cat
https://raw.githubusercontent.com/chromium/chromium/main/components/strings/components_strings_af.xtb
..snip..
https://raw.githubusercontent.com/chromium/chromium/main/components/strings/components_strings_zu.xtb

We could then extend this by piping the output to curl to download the corresponding files:

⇒  gh api /repos/chromium/chromium/contents/components/strings --jq 'map(select(.type == "file" and (.name | test("components_strings_.+\\.xtb"; "i")))) | map(.download_url) | .[]' | (while read url; do curl --silent --remote-name "$url"; done)

⇒  ls
components_strings_af.xtb  components_strings_be.xtb  components_strings_cs.xtb  components_strings_en-GB.xtb   components_strings_fa.xtb     components_strings_gl.xtb  components_strings_hy.xtb  components_strings_ja.xtb  components_strings_ko.xtb  components_strings_mk.xtb  components_strings_my.xtb  components_strings_pa.xtb     components_strings_ru.xtb  components_strings_sr-Latn.xtb  components_strings_te.xtb  components_strings_uz.xtb     components_strings_zu.xtb
components_strings_am.xtb  components_strings_bg.xtb  components_strings_cy.xtb  components_strings_es-419.xtb  components_strings_fi.xtb     components_strings_gu.xtb  components_strings_id.xtb  components_strings_ka.xtb  components_strings_ky.xtb  components_strings_ml.xtb  components_strings_ne.xtb  components_strings_pl.xtb     components_strings_si.xtb  components_strings_sr.xtb       components_strings_th.xtb  components_strings_vi.xtb
components_strings_ar.xtb  components_strings_bn.xtb  components_strings_da.xtb  components_strings_es.xtb      components_strings_fil.xtb    components_strings_hi.xtb  components_strings_is.xtb  components_strings_kk.xtb  components_strings_lo.xtb  components_strings_mn.xtb  components_strings_nl.xtb  components_strings_pt-BR.xtb  components_strings_sk.xtb  components_strings_sv.xtb       components_strings_tr.xtb  components_strings_zh-CN.xtb
components_strings_as.xtb  components_strings_bs.xtb  components_strings_de.xtb  components_strings_et.xtb      components_strings_fr-CA.xtb  components_strings_hr.xtb  components_strings_it.xtb  components_strings_km.xtb  components_strings_lt.xtb  components_strings_mr.xtb  components_strings_no.xtb  components_strings_pt-PT.xtb  components_strings_sl.xtb  components_strings_sw.xtb       components_strings_uk.xtb  components_strings_zh-HK.xtb
components_strings_az.xtb  components_strings_ca.xtb  components_strings_el.xtb  components_strings_eu.xtb      components_strings_fr.xtb     components_strings_hu.xtb  components_strings_iw.xtb  components_strings_kn.xtb  components_strings_lv.xtb  components_strings_ms.xtb  components_strings_or.xtb  components_strings_ro.xtb     components_strings_sq.xtb  components_strings_ta.xtb       components_strings_ur.xtb  components_strings_zh-TW.xtb

Once we have the files locally, we can use the same grep / xpath trick that we used previously to extract all of the translations:

⇒  TRANSLATIONS=`grep --no-filename -r '2354001756790975382' *.xtb` && echo -e "<translations>\n$TRANSLATIONS\n</translations>" | xpath -q -e '//translation/text()' | wc -l
      81

⇒  TRANSLATIONS=`grep --no-filename -r '2354001756790975382' *.xtb` && echo -e "<translations>\n$TRANSLATIONS\n</translations>" | xpath -q -e '//translation/text()' | sort
Alamisho zingine
..snip..
บุ๊กมาร์กอื่นๆ

To save having to download the files again if they haven't changed, we could look at the sha field returned by the GitHub API for each file entry (alongside the download_url field). We could cache the file name, sha, and translation locally; and then would only need to re-download the file if the sha has changed; and read the translation from our cache if the sha is the same as previously.

@0xdevalias
Copy link
Owner Author

0xdevalias commented Dec 5, 2023

We could also add a feature for seeing/restoring recently closed tabs/windows:

See also:

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

1 participant