search

jscher2000 / Firefox-File-Utilities

Mozilla LZ4 File Decryption and Mining Tools
34 stars 7 forks source link

Issue: Stuck on "Processing..." #4

Open Mirrorman95 opened 2 months ago

Mirrorman95 commented 2 months ago

I'm running Windows 10 64-bit. I'm encountering the following issue with the Firefox File Utilities on Mozilla Firefox, but I had the same results with this on Microsoft Edge, too. I recently had a bad Firefox session restore where all of my tabs were blank when they were reopened. This the third time I've had this happen in the past month or so. So, I did what I did the last two times this happened, I backed up the sessionstore-backup files and loaded the ~2.5 MB previous.jsonlz4 into this webpage: https://www.jeffersonscher.com/ffu/scrounger.html I saved the uncompressed JSON files it spit out, and then clicked the Scrounge URLs button. Every time I do this, on Firefox and Edge, it just says "Processing..." for hours with no signs of resolving. The last time I did this, about a month ago with a ~1.6 MB previous.jsonlz4, I didn't have this problem. It gave me a 142 KB html file almost immediately. I went back and tried that older previous.jsonlz4 again, and it said "Processing...done!" almost immediately after I pressed the Scrounge URLs button. Why isn't it working on my newer 2.5 MB previous.jsonlz4?

EDIT: I just tried it with https://www.jeffersonscher.com/ffu/scrounger_20180826.html , and the HTML export works fine in that version.

jscher2000 commented 2 months ago

Sorry to hear that. It probably is an issue with the recursion code. However, once it gets stuck, it doesn't emit any debugging information to provide clues -- unless you notice something in the console (Ctrl+Shift+K).

The "Unstructured URLs" backup option to the right of "Scrounge URLs" probably gives a result similar to the older page because it doesn't have the newer code to parse the back-forward history of each tab to try to understand which is the active URL.

For privacy reasons, it's probably best that I do not receive any files that cause this problem, so unless I encounter it myself, it's going to be hard to debug.

Mirrorman95 commented 2 months ago

Sorry to hear that. It probably is an issue with the recursion code. However, once it gets stuck, it doesn't emit any debugging information to provide clues -- unless you notice something in the console (Ctrl+Shift+K).

The "Unstructured URLs" backup option to the right of "Scrounge URLs" probably gives a result similar to the older page because it doesn't have the newer code to parse the back-forward history of each tab to try to understand which is the active URL.

For privacy reasons, it's probably best that I do not receive any files that cause this problem, so unless I encounter it myself, it's going to be hard to debug.

I appreciate your hasty response. I just wish the older version didn't seem to automatically use "Include Closed Tabs Include Closed Windows Include Back-Forward History". I really don't want any of those to be part of it, because I just want the tabs that closed and not the extra stuff. I guess I'll just have to come up with a regex pattern to remove the "backhist" sections.

EDIT: This RegEx expression I made with help from airegex.pro seems to work for me for removing the tab histories: ^.*<div style="margin-left:2.25em" class="backhist"><span class="bhtoggle" onclick="toggleBH\(this\);">← Back <em>\(earlier pages visited in this tab\)</em>:</span>.*$(?:\r?\n(?!<h4>Tab ).*)*(<h4>Tab .*)?

jscher2000 commented 2 months ago

If you could, hang on to the file that didn't parse. When I get time, I'll create some variations on the script that have a subset of the changes between 2018 and 2024 to see how close I can get to what you're looking for.

Mirrorman95 commented 2 months ago

If you could, hang on to the file that didn't parse. When I get time, I'll create some variations on the script that have a subset of the changes between 2018 and 2024 to see how close I can get to what you're looking for.

I will. I always hang onto those files anyway. But the old version of the website and my regex expression already helped me reopen all of the right tabs, so if you continue to work on this, it would only be to try to improve the Firefox File Utilities.

jscher2000 commented 2 months ago

For testing, I put the embedded session recursion behind a checkbox (off by default) on the regular page, and updated the old logic page so you could omit closed windows/closed tabs/back history:

Mirrorman95 commented 1 month ago

For testing, I put the embedded session recursion behind a checkbox (off by default) on the regular page, and updated the old logic page so you could omit closed windows/closed tabs/back history:

* https://jscher2000.github.io/Firefox-File-Utilities/scrounger.html

* https://jscher2000.github.io/Firefox-File-Utilities/scrounger_old.html

Do you want me to test my lz4 file on them?