Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Status
The first post of this thread is a WikiPost and can be edited by anyone with the appropiate permissions. Your edits will be public.
I was there and didn’t find it


and

 
  • Like
Reactions: TheShortTimer
@Bruninho, thanks for the news (to me) about Browservice.
Do you know if it needs at least Debian 10 to run ? I've tried building it on a Debian 9 virtual machine, and it fails.
 
@Bruninho, thanks for the news (to me) about Browservice.
Do you know if it needs at least Debian 10 to run ? I've tried building it on a Debian 9 virtual machine, and it fails.
According to the developer documentation on github, yes, at least Debian 10. I am using latest Raspbian on Raspberry Pi 3B.
 
Bruninho,

ok, where do I enter this new code ? I know how to compile it with make command, but where do I put all these changes to the code ?


Update: changing the block to an if else and declaring width -= 1; and height -= 2; for the else statement, works. The scrollbars are gone.

I think that probably ALL browsers can use these values? Meaning that an if else is not needed?

and also this - the forward button - also, do I use nano to make these changes ?

To prove or disprove this theory, we need to log all the HTTP requests that Browservice receives. Could you try to add the line
INFO_LOG("HTTP ", request->method(), " ", request->path());

to the beginning of the function Context::eek:nHTTPServerRequest in viceplugins/retrojsvice/src/context.cpp? To apply the change, you can simply re-run the make command and restart Browservice. After this, connecting with IE5 to the modified Browservice instance, see what HTTP GETlines there are in the command line output. If it was running correctly, the output would look something like
topi@laptopi:~/vcs/browservice$ release/bin/browservice 2>&1 | grep HTTP
INFO @ retrojsvice.so src/http.cpp:620 -- Starting HTTP server (listen address: 127.0.0.1:8080)
INFO @ retrojsvice.so src/http.cpp:628 -- HTTP server started successfully
INFO @ retrojsvice.so src/context.cpp:570 -- HTTP GET /
INFO @ retrojsvice.so src/context.cpp:570 -- HTTP GET /1/XirNMFay3yQAXbdv2amTv8tNoKpLoEss/prev/
INFO @ retrojsvice.so src/context.cpp:570 -- HTTP GET /1/XirNMFay3yQAXbdv2amTv8tNoKpLoEss/
INFO @ retrojsvice.so src/context.cpp:570 -- HTTP GET /1/XirNMFay3yQAXbdv2amTv8tNoKpLoEss/next/
INFO @ retrojsvice.so src/context.cpp:570 -- HTTP GET /1/XirNMFay3yQAXbdv2amTv8tNoKpLoEss/
INFO @ retrojsvice.so src/context.cpp:570 -- HTTP GET /1/XirNMFay3yQAXbdv2amTv8tNoKpLoEss/image/1/1/1/1500/886/0/
INFO @ retrojsvice.so src/context.cpp:570 -- HTTP GET /1/XirNMFay3yQAXbdv2amTv8tNoKpLoEss/image/1/2/0/1500/886/0/
<Continues with lots of image requests>

However, if my theory is correct, you only get the first four HTTP GET requests, after which IE5 uses the cached pages.
An easy way to get around this bug is to disable the native Forward button forwarding by changing the initial value of preMainVisited_ from false to true on line 63 in viceplugins/retrojsvice/src/window.cpp.
 
Btw, Netscape 7.0.2 is a No go, Icab for OS 9 too is a no-go, including opera + Wamcom and Mozilla.. just for kicks, I tried Browservice with LWK in Leopard - sadly, it too did not work.
 
Btw, Netscape 7.0.2 is a No go, Icab for OS 9 too is a no-go, including opera + Wamcom and Mozilla.. just for kicks, I tried Browservice with LWK in Leopard - sadly, it too did not work.

finally, where do I find this, so I can add it ?



To prove or disprove this theory, we need to log all the HTTP requests that Browservice receives. Could you try to add the line

INFO_LOG("HTTP ", request->method(), " ", request->path());

to the beginning of the function Context::eek:nHTTPServerRequest in viceplugins/retrojsvice/src/context.cpp? To apply the change, you can simply re-run the make command and restart Browservice.

in other words, where would I add it - in the front of that line or below it ?
 
You don’t need to mess with it if you recompile browservice with the changes he proposed. he detailed them on the issues page of his github project page. have a look on them.
Oh, ok.. so are they in an updated version of Browservice ? I have 0.9.2.8 I think.
 
you are saying i don't need to do anything ? Where do I put this ?

INFO_LOG("HTTP ", request->method(), " ", request->path()); - do i put this under the function on

viceplugins/retrojsvice/src/context.cpp


void Context::eek:nHTTPServerRequest(shared_ptr<HTTPRequest> request) {

REQUIRE_API_THREAD();
REQUIRE(state_ == Running);
 
Added in a proper supported hardware list to the first post. Interestingly, the support page linked before was missing a few models. Even Apple themselves couldn't keep track of all of the models. 😆
 
  • Like
Reactions: Bruninho
Thank you! I had a hard time trying to make a list too, and I was asleep halfway through the original list from the support page 🤣 so I just linked it there instead. BTW, how is the Mac Mini M1 for you? Considering one for my dad to convince him to make the jump from his 2011 Intel Mac Mini to a M1 Mac Mini :p
 
Bruninho,

ok, where do I enter this new code ? I know how to compile it with make command, but where do I put all these changes to the code ?


Update: changing the block to an if else and declaring width -= 1; and height -= 2; for the else statement, works. The scrollbars are gone.

I think that probably ALL browsers can use these values? Meaning that an if else is not needed?

and also this - the forward button - also, do I use nano to make these changes ?

An easy way to get around this bug is to disable the native Forward button forwarding by changing the initial value of preMainVisited_ from false to true on line 63 in viceplugins/retrojsvice/src/window.cpp.

The developer already explained there on github, in your post excerpt above. You need to be more patient and carefully read his documentation and Issues thread on github. I highlighted there for you. Impossible to be more clear than that. 🤷‍♂️
 
The developer already explained there on github, in your post excerpt above. You need to be more patient and carefully read his documentation and Issues thread on github. I highlighted there for you. Impossible to be more clear than that. 🤷‍♂️
No no... I know, I saw what .cpp its in, just need to know under the void function, is that where I enter it ? I read it, but when I clicked on the .cpp file affected, I saw the void function and a couple of other things underneath. I was not sure if this "INFO_LOG("HTTP ", request->method(), " ", request->path());" goes like this:

code from .cpp viceplugins/retrojsvice/src/context.cpp? -

void Context::eek:nHTTPServerRequest(shared_ptr<HTTPRequest> request) {

REQUIRE_API_THREAD();
REQUIRE(state_ == Running);

"Then enter here-->" INFO_LOG("HTTP ", request->method(), " ", request->path());

Putting it all together:

void Context::eek:nHTTPServerRequest(shared_ptr<HTTPRequest> request) {
REQUIRE_API_THREAD();
NFO_LOG("HTTP ", request->method(), " ", request->path());

Thats all...
 
And I am now off to my bed after searching for a few of my favorite childhood select games versions for mac: SimTower, Carmen Sandiego... I have all the DOS and Windows 9x games, the most recent acquisition was Moto Racer 2 (Windows). Gotta find Monster Truck Madness for Win 9x now. And a racing game for OS9, because I happen to be a real-life indoor karting racer, too...

[sound on] *Where, In The World, Is Carmen Sandiego...* [/sound off]
 
No no... I know, I saw what .cpp its in, just need to know under the void function, is that where I enter it ? I read it, but when I clicked on the .cpp file affected, I saw the void function and a couple of other things underneath. I was not sure if this "INFO_LOG("HTTP ", request->method(), " ", request->path());" goes like this:

code from .cpp viceplugins/retrojsvice/src/context.cpp? -

void Context::eek:nHTTPServerRequest(shared_ptr<HTTPRequest> request) {

REQUIRE_API_THREAD();
REQUIRE(state_ == Running);

"Then enter here-->" INFO_LOG("HTTP ", request->method(), " ", request->path());

Putting it all together:

void Context::eek:nHTTPServerRequest(shared_ptr<HTTPRequest> request) {
REQUIRE_API_THREAD();
NFO_LOG("HTTP ", request->method(), " ", request->path());

Thats all...

Well, ask him on github. He’s the developer. I already told you that I am not a programmer. I just did as he instructed crystal clear there: Read again the line I highlighted for you in the previous post.
 
  • Like
Reactions: Amethyst1
Tried it also on Netscape, icab, opera - NO GO.. seems to only work on Classila, Mozilla browser, and IE 4, 5, 5.1.7 - may try safari.
 
There is no version of this game for Mac.

Javascript is the reason why Netscape and Opera for OS9 do not work with Browservice. It works on Classilla because Cameron Kaiser did a fine job in giving a bit of an improvement there. But then, Browservice still works with a way older IE on Windows 3.11, so I can't say for sure. Unlike WRP, Browservice is heavily based on javascript to plot a toolbar/viewport to navigate.

I can't speak for iCab, since I have never managed to make it work even with native pages, maybe some settings for javascript there has to be enabled or disabled.

I've set Browservice on IE 5.1.7, I hide the IE address bar so I can force myself to use Browservice's one, I've set the browservice URL address as the default home address, so whenever I want to (re)start my navigation flow, I hit the home button. For the sole purpose of browsing more modern pages, I use IE with Browservice on OS 9. Since there is no way to store them as favorite bookmarks, I might have to write a simple html page with the links for them to be accessible through Browservice.

For other purposes, I use Classilla with native accessible web pages when I have to download something or do not need Browservice at all. Pages like Macintosh Garden and macOS9lives. I think that the current setup with both browsers is a perfect solution for the time being.
 
I agree and I figured out where to place that line of code - under the void function. I also set the other one to say true, instead of false. I will recompile sometime this week. Now, if we can have the developer get some kind of sound output, this will be a great solution to browsing web on OS 9. In the meantime, my PowerBook G4 has enough power to compile Classila with the crypto Ancienne for TLS 1.2 - that is my next project.
 
I’ve read that certain graphics upgrades for G3s (PCI) and G4s (AGP) aren’t supported by OS 9, and only work for OS X.

If that’s true, does anyone know the best graphics cards that still retain OS 9 support?
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.