Subaru Forester Owners Forum banner
  • The "Garage" feature is for images of YOUR VEHICLE/S only - no blanks or other unrelated images please, thanks
141 - 160 of 186 Posts
Finally, this spring break I am going to change my diff oil in my 2019 Sport. Tried to get the high school auto shop team to do it, but they are busy and right now they are first semester students. Got everything and have been watching videos.

Got red line, but then have to get a pump to get the oil in since it is kinda crowded. So ordered the slippery Pete with the adapter I can work on another bottle since the slippery Pete does not fit the Red Line quart oil.

Then saw that I needed crush washers.

Then got to thinking I really should get a manual.
Saw the STIS and can get 3 days for 35.

Has anyone done this? Can you get it and download it as pdf? Heard earlier years were all html, but they might have changed to pdf.

If anyone has done the diff fluid on a 2019 and has pictures and advice - thanks in advance.
 
Has anyone tried to pay the $35 for 3 days access to view their service manual? I am curious because I'd like to get my '22 Forester's service manual but I in downloadable form since using the aforementioned site would be cost prohibitive as a DYIer.
 
@Javi Per the rules around here....
7. We do not allow the distribution, or offers of provision of copyrighted material, such as links to illegal download sites for service manuals, music, pirated software, etc.

The last thing this great resource needs is a lawsuit from SOA.
 
I created a Chrome Extension that adds a "Section" link to the "Print" button in the STIS HTML service manual. This allows users to print the full section of the service manual instead of each individual chapter. It works by automating the process of loading each individual "chapter" on to the same page.


I no longer have access to the STIS website since I only paid for the 3-day subscription and only tested this on the '19 Forester service manual, so no guarantees this works for other cars. But thought it could be useful for others as well. If you are more computer savvy you can find the code here: GitHub - schultetwin1/STISSectionPrinter: A Chrome extension to make it easier to print the full section of a service manual for your Subaru from STIS Website
 
I created a Chrome Extension that adds a "Section" link to the "Print" button in the STIS HTML service manual. This allows users to print the full section of the service manual instead of each individual chapter. It works by automating the process of loading each individual "chapter" on to the same page.
I've never logged into the STIS. What does it look like to use this button to print the sections? Is it like, buy the 3 days, use this button 10-20 times to print the sections to PDF (I don't know how many sections there are) and then that's it? Would doing this cause someone to hit the per-day file limit?
 
Good question for @schultetwin1 on how his extension counts towards the file limit.

I saw nothing about a per day limit, but an earlier post noted a 50/hour limit.

I downloaded my manual before the chrome extension existed, and possibly before a per/hour limit existed. Along with tech bulletins and other miscellaneous files, it totaled around 200 files and was slow and laborious.
 
Is it like, buy the 3 days, use this button 10-20 times to print the sections to PDF (I don't know how many sections there are) and then that's it? Would doing this cause someone to hit the per-day file limit?
That's exactly right, but I think it's more like 30 sections. The per-day file limit is a good question. I don't remember hitting any such limit when I was developing the extension but I didn't do all the downloads in 1 day so I'm not sure if you would run into it or now.
 
Hey @schultetwin1 I purchased the 72-hour access yesterday with hopes of using your extension to download all the sections, so the clock is ticking for me. I'm only just starting with the first section "General Description", and right away I see some differences in the output between what your section printer extension generates and using the STIS "print chapter" button.

Could this be because the code in the service_worker script in your extension will only work for the same model 2019 Forester as yours, since it includes, on Line 7,
..."/stis/doc/htmlDiagnostics/19_FORESTER_G8240BE_V42/contents/data/"... ?

I have no prior experience with building browser extensions, or editing and re-compiling the source code like you have available on github.
  • Can you tell me what I would need to change / what I would need to do in order to edit your script to pull the pages for a 2021 Forester?
  • (I imagine there is underlying code on the STIS site that references their internal directories/folders corresponding to specific models, to pull up the correct files for 'printing'.)
    • When I open links within pages, or use the STIS Print Chapter button, the link address (and result in browser address bar) is in this format: //techinfo.subaru.com/stis/diag/2021/Forester/contents/data/print/
    • (I don't know if that is basically the "location" that would effectively replace the "//techinfo.subaru.com/stis/doc/htmlDiagnostics/19_FORESTER_G8240BE_V42/contents/data/print/" in your code? Or if not, how to determine what it should be?)
I'd be very grateful if you could assist! I really don't want to lose out on the time I have left to download the material. THANKS !
 
You are 100% correct. I hard coded it to work with only 2019 Forester on accident. @techie-howie messaged me separately. He told me he was able to figure out how to edit the Chrome extension to get it to work for him. I also have fixed the Chrome extension on my end and submitted the fix for review in the Chrome store. Once approved by Chrome, it should be live for everyone. The fix is in version 0.2 (currently version 0.1 is what is released).

@techie-howie also mentioned that when loading some of the larger sections it slowed his computer down to be basically unusable. I had the same problem. With some of the larger sections I would have to walk away from the computer for 20-30 mins for before I could use it again. And I have a newer MacBook. So those will older machines might not be able to get that extension to work.
 
Hi, all Subieowners,

STIS - offline interactive service manual, updated high level guide


I found myself turning some nuts and bolts more and more often on my subie so I decided to get a service manual for the car. Inspired by this post, I purchased access to STIS and to my surprise found out that it no longer works as there are no more PDFs to be grabbed, just an online system. How smart - if you want to check the service guide, you have to pay every time...

I will not be including any materials here for obvious reasons but will share my thought process on how one might be able to obtain a working offline copy of the same manual.
This is not a full step by step guide, some level of technical understanding is required to complete it.

To view online version of the service manual go to STIS, Service Diagnostics -> Year -> Model. It will then open in a new tab for you.

Site cloning software
This was the first thing that I tried. But the site navigation is as simple as it is not cloning friendly. Side menus are generated dynamically by JS, do not contain a href element, and content of the site is displayed in an iframe. Thus any cloner without JS support will not work. I found myself using Cyotek Web copy. It is, however, not flexible/smart enough to click on the links that do not look like links - and that's exactly what this site has. I was successful in cloning only the home page. Not very useful, however I did keep this copy to see if I can semi-manually fill in the blanks.

JS generated menu and links to the content.
While debugging site navigation I found a file called menus.js. This file contained the information for the dynamic links to the content itself.
The folder structure is as follows: /data/{4 character chapter identifier}/{same 4 characters+extended file name}.html
a simple grep over this file generated the list of the file IDs that I would need to download.

A little curl magic allowed me to grab all 5800 files in about 1 hour. No throttling or download limit was applied - maybe because these are web pages and not files.

Basically, I used a curl request generated by Chrome for one of the links I accessed to make sure I looked as much like a browser as possible. You'll need to supply your own cookies, though.

Bash:
while IFS= read -r line; do
    encoded_line=$(printf %s "$line" | tr -d '\r')
    url="https://techinfo.subaru.com/stis/doc/htmlDiagnostics/23_FORESTER_G8330BE_V5/contents/data/$encoded_line"
    echo "Downloading: $url"
    curl --url "$url" \
      --create-dirs \
      -o "data/$encoded_line" \
      -H 'Accept: */*' \
      -H 'Accept-Language: en-US,en' \
      -H 'Cache-Control: no-cache' \
      -H 'Connection: keep-alive' \
      -H $'Cookie: !!!!!your cookies!!!!' \
      -H 'DNT: 1' \
      -H 'Pragma: no-cache' \
      -H 'Referer: https://techinfo.subaru.com/stis/doc/htmlDiagnostics/23_FORESTER_G8330BE_V5/contents/index.html' \
      -H 'Sec-Fetch-Dest: script' \
      -H 'Sec-Fetch-Mode: no-cors' \
      -H 'Sec-Fetch-Site: same-origin' \
      -H 'User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36' \
      -H 'sec-ch-ua: "Not_A Brand";v="8", "Chromium";v="120", "Google Chrome";v="120"' \
      -H 'sec-ch-ua-mobile: ?0' \
      -H 'sec-ch-ua-platform: "Windows"' \
      --compressed;
done <subielinks

Example lines from the subielinks file:
s068/s06852102z.html
s068/s068521030.html
s068/s068521da5.html
s068/s068521da6.html

At this point, I thought that images are embedded via data: URI scheme, but it turned out most of them are not.

Pulling the images/schematics
After analyzing a couple of files and understanding that the images actually share a common folder and with some grep magic later:
grep -hroP "grph_ent/.*?\.svg" . | uniq > grph_ent_list
Keep in mind that I needed to use non-greedy matches and thus reverted to perl syntax. Since perl installation can vary from platform to platform, the above command might not work.
I was able to extract all the references to the images.
I fed this list of 5800 images to the same script I used before.

I ended up pulling some additional CSS files that were missed by the cloning software.

Result - a local working copy of an interactive service manual.
I bet someone can create a recipe for all of these actions but I'm too lazy for that and my 3 days that I paid for are almost done. If you do - please share it forward:)
 
141 - 160 of 186 Posts
You have insufficient privileges to reply here.