Coding Projects and Duga Heaven :)

« first day (169 days earlier) ← previous day next day → last day (3691 days later) »

Duga

00:00

RELOAD!

[skiwi2/HearthMonitor-LogReader] 1 commits. 0 issues opened and 0 closed

[skiwi2/GithubHookSEChatService] 0 commits. 0 issues opened and 1 closed

12 hours later…

skiwi

12:27

hey

1 hour later…

Marc-Andre

13:49

hey @skiwi!

skiwi

hey @Marc-Andre

Weird Java issues here..

Take this code

        Thread thread = new Thread(() -> {
            try (CloseableLogReader logReader = new MonitoringFileLogReader(Files.newBufferedReader(logFile, StandardCharsets.UTF_8), new ABCEntryParsers())) {

                assertTrue(logReader.hasNextEntry());
                logReader.readNextEntry();

                assertTrue(logReader.hasNextEntry());
                logReader.readNextEntry();

                assertTrue(logReader.hasNextEntry());
                logReader.readNextEntry();

                //still believes there is a next entry

Somewhere in the try statement it throws an ClosedByInterruptException, but it uses the general Exception catch block

Marc-Andre

humm in try-with-resources if an Exception happen in the try itself I think there is a special behavior

skiwi

Apparently it's an UncheckedIOException, ugh

Marc-Andre

Suppressed Exceptions

An exception can be thrown from the block of code associated with the try-with-resources statement. In the example writeToFileZipFileContents, an exception can be thrown from the try block, and up to two exceptions can be thrown from the try-with-resources statement when it tries to close the ZipFile and BufferedWriter objects. If an exception is thrown from the try block and one or more exceptions are thrown from the try-with-resources statement, then those exceptions thrown from the try-with-resources statement are suppressed, and the exception thrown by the block i…

skiwi

printing ex.getMessage() wasn't a good idea

Marc-Andre

13:54

there that too

docs.oracle.com/javase/tutorial/essential/exceptions/…

skiwi

Those suppressed exceptions are also ugly for sure

I was close to having it correct

but errors on another thread didn't propogate yet

Duga

14:12

test

[skiwi2/HearthMonitor-LogReader] skiwi2 pushed commit 2a147f9c to develop

> Refactored the LogReader interface:
- The readEntry() method has been changed to the readNextEntry() method.
- The readNextEntry() method no longer throws a NoMoreInputException.
- A hasNextEntry() method has been added.
- The AbstractLogReader class now takes an Iterator<String> rather than overriding a method that supplies a String.

[skiwi2/HearthMonitor-LogReader] build #36 for commit 2a147f9c on branch develop passed

1 hour later…

skiwi

15:39

This is... salty

I was on 4 hp for a very long time, and got him down to here from like 30

Duga

16:28

[skiwi2/GithubHookSEChatService] Vogel612 commented on issue #47: Nothing gets posted to the chat anymore once it went down

> Looks like the Responsible thread terminating is the cause of this. It could be remediated by restarting the thread, given it terminates. It might be sufficient to alternatively wrap the executed code into a pokémon-catch and selectively rethrow

[skiwi2/GithubHookSEChatService] Vogel612 commented on issue #34: Context-aware line breaking

> Ticket on JavaBot has been closed as completed. Linebreaks are not given preference over there, mostly since we're not expecting linebreaks :( Feel free to draw inspiration from our implementation ;)

Vogel612

Hey @SimonAndréForsberg You wanted to shoot some tickets at JavaBot?

Simon André Forsberg

16:48

@Vogel612 yes, I'm not sure if I perhaps should wait for JCE though

Vogel612

I doubt it's going to change the underlying problems..

Simon André Forsberg

might be true

so I guess I'd better make issues of what I have to say

@Vogel612 I see this in a comment inside the code, but is there a github issue for it? // TODO: get the chatInterface solved via Dependency Injection instead.

Vogel612

no there's no issue yet..

also that should be a fixme

but I fear it will have a rattail of other stuff.

Simon André Forsberg

I was considering trying to work on that

and I probably would have, if I could have logged in on chat.stackexchange.com

currently, logging in is the big block for me. unfortunately, I didn't have much time last time to work on that.

Vogel612

you want a place to test it??

Duga

17:00

[skiwi2/GithubHookSEChatService] Zomis commented on issue #47: Nothing gets posted to the chat anymore once it went down

> @Vogel612 from what I have noticed so far, there is no real place to put a pokemon-catch. The logging shown above is from this method:

@ExceptionHandler(Exception.class)
@ResponseStatus(HttpStatus.BAD_REQUEST)
public void handleException(final Exception ex, final HttpServletRequest request) {
LOGGER.log(Level.SEVERE, "exception", ex);
}

i.e. there is no "rethrowing" going on that I have been able to locate.

Simon André Forsberg

@Vogel612 I have plenty of chat rooms already that I could use... wait, are you talking about your repo? I haven't understood completely what your repo does, or how it could help

Vogel612

well it's for offline testing.

no chatroom

just you and junior

Simon André Forsberg

ok, so I guess you could call it a chatroom simulator?

Vogel612

kinda

Simon André Forsberg

Vogel612

17:12

but yes, that chatInterface thing has been bugging me for a long time already

or rather the whole setup thing...

since there is code written and hand tuned to wire things up

skiwi

17:45

Is there a way I can "measure" how extensive my refactoring has been?

Vogel612

deletions....

diffstat

skiwi

Let's see how I set that up

Duga

[skiwi2/HearthMonitor-LogReader] skiwi2 starred us

[skiwi2/HearthMonitor-LogAPI] skiwi2 starred us

[skiwi2/HearthMonitor-AndroidApp] skiwi2 starred us

skiwi

Still don't get why it doesn't just auto-star my own repos ^

github.com/skiwi2/HearthMonitor-LogReader/compare/…

> Showing 37 changed files with 1,295 additions and 1,403 deletions.

Vogel612

holy..

skiwi

17:49

That's a lot?

I think the CreateGameEntryParser.java is one of the most important ones

It only needed all those other changes to work

Vogel612

"only"...

and all don't go all that well together usually..

also 1.2k adds and 1.4k dels are a lot for one thing..

skiwi

The only thing that's bugging me right now is that my AbstractLogReader class has no more abstract methods anymore, but it's a class that is supposed to be extended

Vogel612

make it "DefaultLogReader" then...

skiwi

I could give it a private constructor, but it would still show up as an instantiable class then

Simon André Forsberg

18:15

@skiwi how would you plan to extend it if it has a private constructor?

skiwi

@SimonAndréForsberg Good point, it needs a protected non-default constructor

I'm mostly bugged with needing to rename it from AbstractLogReader to something else

Simon André Forsberg

@Vogel612 yes, that is what is bugging me as well. Program should really be injected with the commands. Or expose the underlying bot so that it can have no commands at first, and those commands are added to the bot from another class

@skiwi then why did you refactor it in the first place?

skiwi

Because it needed to be refactored

Simon André Forsberg

19:05

^^ good reason

@Vogel612 I was wondering about this regex: "([*_]{1,3}.*?[*_]{1,3})"

Vogel612

italics, bold and bold-italics

Simon André Forsberg

also I need to test

huh, didn't know _ worked for bold as well

Vogel612

it ~~should~~ does

Simon André Forsberg

but what about the .*? ? is the ? really necessary there?

Vogel612

non-greedy matching

else we might consume multiple tokens

Simon André Forsberg

19:06

ah, ok.

Vogel612

__if you think__ about somethin _like this_

actually that should be written differently..

Simon André Forsberg

I would have normally done [^*_]* in this case (excluding _ and *)

Vogel612

"(([*_]{1,3}).*?(\\1))"

Simon André Forsberg

but your way might be better, didn't know that even worked...

Vogel612

this one up there is bactracking to have the ending group match the starting group

Simon André Forsberg

19:08

Vogel612

but in fact the tokenizing fails for cool stuff like:

__some_ cool markdown_

_some cool markdown_

^^ this is rendered differently in chat and questions..

Simon André Forsberg

How about [links to someplace](yadayada) ? Those aren't currently included in the current regex...

Vogel612

they are in another...

Simon André Forsberg

@Vogel612 possibly sounds like a bug for MSE. but that's quite interesting

Vogel612

private static final String  linkTokenRegex          = "(\\[[^]]++\\]\\(https?+:\\/\\/[^\\s\"]++(\\h++\"[^\"]++\")?\\))";

Simon André Forsberg

19:10

ah, alright

Vogel612

though I must admit that the link token is buggy.

Simon André Forsberg

I can imagine that. getting a link right is hard...

what's buggy with it?

Vogel612

it doesn't work properly with title text.

Simon André Forsberg

[like this?](http://example.com)

Vogel612

there's two test-cases for it..

but I remember some corner case being borked.

1 hour later…

Duga

20:20

Fine by me too

2 hours later…

Phrancis

21:52

Monking

Simon André Forsberg

hey @Phrancis

Phrancis

22:27

@SimonAndréForsberg I went ahead and sent them a message to see if there was any interest, will let you know.

Simon André Forsberg

ok, thanks

« first day (169 days earlier) ← previous day next day → last day (3691 days later) »

all rooms

Transcript for

Jan17

Jan '1518

Jan19

Coding Projects and Duga Heaven :)

Minesweeper Flags, Zomis' games, PDB, MTG, Cardshifter, and al...

browser-tabs duga factorio fleet-command kotlin minesweeper pdb pineapples vue.js

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC