[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Fixed in DC++ 0.800. ** Changed in: dcplusplus Status: Fix Committed => Fix Released -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title: Plugins support Status in DC++: Fix Rele

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Changed in: dcplusplus Status: In Progress => Fix Committed -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title: Plugins support Status in DC++: Fix Committed Bug description:

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Oh one point I forgot to cover... missing settings types, tbh I'm not sure if I want to mimic dcpp here... as in why floats over doubles (yes floats could be technically faster, on the x86 architecture, but if you want precision for math then double should be the way to go) for example... or why no

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Regarding thread safety, being something arne brought up... the result is to sum it up in one word: ugly. The two major functional interfaces (DCHub and DCQueue) now have release(...) and copy(...) functions in them, the respective *Data structs also have an isManaged member that is False if releas

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Ok, few comments in response to the points brought up in dcdev chat: 1. Regarding the types of id'ing used... and the differentiation between different groups. The meaning between generic vs dcpp is very much obvious as stated; other than that the ids can be described as: [generic|dcpp].[someth

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Any updates? considering the last 10 months of updates to the DC++ it would be nice to know the status of any progress or an updated patch if possible Crise. -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

So far mainly in my head, since various things have kept me busy... that said I do agree with your comments for the most part. -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title: Plugins s

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Any updates? -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title: Plugins support Status in DC++: In Progress Bug description: Ok, I'll leave out the sales pitch... here is a potentia

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Branch linked: lp:~markuwil/dcplusplus/dcpp-plugins -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title: Plugins support Status in DC++: In Progress Bug description: Ok, I'll leave

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

http://lcsd05.cs.tamu.edu/slides/keynote.pdf has a nice presentation on this topic, entitled "How to Design a Good API and Why it Matters" by Joshua Bloch. It points out some specific hazards and advantages. Some of the more interesting suggestions: -It's probably useful to have three distinct plu

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Well a branch has been created by Crise for plugins support, just waiting for him to finish what he's doing in regards to changes with the API here's the branch adress https://code.launchpad.net/~markuwil/dcplusplus/dcpp-plugins -- You received this bug notification because you are a member of

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

+1 for branch... -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title: Plugins support Status in DC++: In Progress Bug description: Ok, I'll leave out the sales pitch... here is a pote

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I've brought this up once before but not sure if anyone noticed it. With the latest developments of a plugin page, this seems even more relevant: Wouldn't it make sense to track the code changes for this bug in a separate branch? Patches are great for small changes but become cumbersome when intro

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Patch removed: "plugin page final" https://bugs.launchpad.net/dcplusplus/+bug/587597/+attachment/1784643/+files/pluginPage.patch -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title:

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

will do it was requested to be added in here but no problem -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title: Plugins support Status in DC++: In Progress Bug description: Ok, I'll

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

iceman, could you open a separate bug for the gui...? -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title: Plugins support Status in DC++: In Progress Bug description: Ok, I'll leave

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Well finally completed the plugin page, everything is fully functional =) ** Patch removed: "New PluginPage patch" https://bugs.launchpad.net/dcplusplus/+bug/587597/+attachment/1784492/+files/pluginPage.patch ** Patch added: "final version" https://bugs.launchpad.net/dcplusplus/+bug/58759

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Updating my original patch to a somewhat more working one enjoy =) (will try to update asap) ** Patch removed: "pluginPage.diff" https://bugs.launchpad.net/dcplusplus/+bug/587597/+attachment/1784053/+files/pluginPage.diff ** Patch added: "New PluginPage patch" https://bugs.launchpad.net/d

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Ok, so out of pure interest I decided to start making a plugin page for settings, although it's very imcomplete and needs a bit of help id figure id post the diff here and maybe get some of the things finished off to work properly my biggest problem currently is adding entries to the list, the add

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Attachment removed: "DCPlusPlus.exe" https://bugs.launchpad.net/dcplusplus/+bug/587597/+attachment/1730662/+files/DCPlusPlus.exe -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title:

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Changed in: dcplusplus Status: New => In Progress -- You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. https://bugs.launchpad.net/bugs/587597 Title: Plugins support Status in DC++: In Progress Bug description: Ok, I'll

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

For now, I'd say it's ok to call the pluginmanager directly - eventually it is probably the Speaker that should be extended so that an observer can halt further processing of the event in which case the pluginmanager would use that to suspend execution. I agree that an observer-based solution would

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

So, to address the issue of PluginManager::onFooEvent calls that Steven brought up... to do that the existing listeners in dcpp are useless since most of them are fired in "past tense" in other words fired after an action is complete. So even if it is possible to address the listeners not being ab

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Updated patch, for you lazy people not bothered to fix 3 extremely simple conflicts... no other changes (haven't had the time unfortunately) ** Patch added: "plugins-new.patch" https://bugs.launchpad.net/dcplusplus/+bug/587597/+attachment/1756644/+files/plugins-new.patch ** Patch removed: "pl

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Why add a compiled DC++ here ? seems kinda redundant since we all are able to compile it ourself -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC++: New Bug descrip

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Attachment added: "DCPlusPlus.exe" https://bugs.launchpad.net/dcplusplus/+bug/587597/+attachment/1730662/+files/DCPlusPlus.exe -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to D

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I'll first answer about the dcptr_t objects, plugins don't use them... think of them as "handles". They could even be obfuscated instead of being direct pointers to dcpp objects. The object members of the structs are simply something the party which implements callbacks use (unless otherwise state

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Personally, I would prefer a less powerful but cleaner and safer plugin system that could not affect what happens after the plugin processes the event. Cleaner because it could use the listener system for all communication. Safer because it would be read only data passed and could be processed in a

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I had a goal of keeping it pure C API... which allows it, in theory, to be used with languages such as delphi and (yack) VB. If this is doable as C++ only extension, without breaking the underlying C API then it is certainly worth it I'd say... although it would be strictly tied to dcpp based plug

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I have one thing. What about putting Client abstract class as a part of plugin API? Then new procotol could be added by plugins (for example DHT could be a separate part). -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcpl

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I agree, but as explained earlier plugins need to be able to affect what happens after plugin has processed an event... which is why using the existing listeners is not the ideal solution in 9 cases out of 10... to be completely honest I even considered not using any listeners, except timermanager

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

just a nitpick: the current event driven architecture (Speaker class) is not asynchronous; that is the reason GUIs must try not to do any lentghy operations when they process these events. still, using it would be nice. -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this b

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

The calls to pluginmanager return if they are true, because that implies plugin handled that event for us, ie. we don't want core doing anything about that particular instance of it anymore. I agree that the plugins should not be given the possibility to put the core into an unsafe state, however,

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Jacek hasn't reviewed this but we requested it several times over the last year so for now we rely on your reviews and adjust to em and we have internal discussions in dcdev over how it works so for now it has to do with keeping atleast this discussion up most of the core developers are on board al

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Looks like most of the things I mentioned are fixed. I still have some major issues with it though. - Can you explain why the calls to the PluginManager will return if they are false? For example: if(PluginManager::getInstance()->onIncomingConnectionData(this, aLine)) return; in dcpp/UserConnectio

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Changed in: dcplusplus Status: Opinion => New -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC++: New Bug description: Ok, I'll leave out the sales pitch.

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Updated patch, still doesn't have all the changes I had wanted to include but at the same time I realised months had passed. The attached patch also has one line change to RichTextBox.cpp which is unrelated to plugins but I was unable to compile under mingw without it (oddly enough QueueFrame comp

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Er. Lua API as presented by Lua, that is (it doesn't propose any particular changes that I can tell to the C side). -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC+

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

For the record: Thor's Lua API proposal lives at . -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC++: Opinion Bug description: Ok,

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Changed in: dcplusplus Status: New => Opinion -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC++: Opinion Bug description: Ok, I'll leave out the sales pi

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Pardon me, but I was not arguing against DC++ naming standards, I was simply stating my reasons for naming them the way I did. Besides (if it is decided that this gets into DC++) you don't need me to change some capitalisations, anyone can do that since I won't be the one putting it in anyways as I

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

On Tue, 2010-06-01 at 09:08 +, Crise wrote: > The functions which don't follow camel-case are not for the core to use, > but for plugins, well one of them is called from MainWindow but that's > an exception because the hook in question does not need any preparatory > work so it would be a waste

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

This is the same patch as before, but converted from LF to CR/LF formats. The GnuWin32 patch 2.5.9 version seems to prefer it this way. ** Patch added: "plugins.patch" http://launchpadlibrarian.net/50274613/plugins.patch -- Plugins support https://bugs.launchpad.net/bugs/587597 You received t

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Also as cologic stated above... this limited exposure is intentional (I hate not being able to edit my comments), not only because the other path would be significantly more complex to create while keeping it a pure C api but also because this way, as stated, have more precise control over what plu

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Short answer: not directly... Long answer: It doesn't support them right now directly... because the listeners use so many C++ specific features it was easier just to expose select listeners through hooks also it allows more precise control over what info plugin has access to. There is a way to a

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Ah, I should re-state that comment, since the phrasing got confused: "It's not a complete 1:1 mapping onto the current DC++ internals. Instead, it maps onto a slightly more abstracted set of useful DC semantics. This is intentional." -- Plugins support https://bugs.launchpad.net/bugs/587597 You r

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I certainly hope not. (The answer is no, in general, with the current patch. It's not a complete 1:1 mapping between DC++ internals and a somewhat more abstracted set of DC semantics, nor should it be.) -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Does this plugin mechanism support current DC++ listeners? I mean that if something in core will call fire(something,a,b,c), plugins will be able to execute its code for such event? -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a memb

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I'm willing to assign copyright on the bits of BCDC++ I see in there (e.g. the SettingsManager changes) to arne, yes. -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in D

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Oh regarding the copyright, well for the pieces that come directly from bcdcpp that's the place you need to ask... as for the rest I wouldn't have made this patch unless I was ready to give over the copyright. But if you need me to say/write it: The copyright is all yours arne... -- Plugins supp

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

The functions which don't follow camel-case are not for the core to use, but for plugins, well one of them is called from MainWindow but that's an exception because the hook in question does not need any preparatory work so it would be a waste to make another on* function in pluginmanager for it (S

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

all fixable tnx steven for the feedback -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC++: New Bug description: Ok, I'll leave out the sales pitch... here is a pot

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

- PluginManager::onChatDisaplay - grammar - New files show copyright arne, but unless the original patch writers explicitly give copyright to arne it should be copyright the original authors. As I understand, DC++ requires original patch writers to hand over copyright. - It seems strange to me th

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Patch removed: "patch that you can actually apply" http://launchpadlibrarian.net/49406269/plugins.patch -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC++: New

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Here is an updated patch spotted a stupid mistake I made in AspectChat... ** Patch added: "plugins.patch" http://launchpadlibrarian.net/49447956/plugins.patch -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-te

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Both approaches are good... though in a sense since bcdc still uses 5.1 the latter option might be better to start with, so users can just copy over their scripts from bcdc without changes (this would help in having users move from bcdc to DC++). Also rember that it's not necessarily practical to

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Sure, for the moment it makes more sense to stay on 5.1. But 5.2 might be close enough to release that by the time this is committed to DC++ (if it is), then the the initial Lua standard for DC++ to use could be 5.2; alternatively, one could start with 5.1 and just specify that anything that'll bre

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I haven't upgraded to 5.2 lua since it's not marked as stable yet and because of the whole "and everything can still change" thing. -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I haven't tried to apply/compile that patch yet, but will do so. Also, worth being aware that Lua 5.2 is coming along - see http://www.corsix.org/content/look-lua-52-work3 for changes. Shouldn't have a major effect, but probably the Lua implementation should be 5.2-based rather than 5.1-based. --

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Here is a fixed patch, since the initial one seems to have ended up as corrupted somehow... ** Patch added: "patch that you can actually apply" http://launchpadlibrarian.net/49406269/plugins.patch -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification bec

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Patch removed: "against rev 2155" http://launchpadlibrarian.net/49398711/plugins.patch -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC++: New Bug description

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

You could actually apply the patch? when I toast tried it and then later me all we got was patch exe crashing... also just noticed that there is something funny going on with line endings, I thought I made sure all were the unix flavour. -- Plugins support https://bugs.launchpad.net/bugs/587597 Y

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I haven't yet looked in detail at this revision of the code, so I can't comment on it, but at a glance it looks similar to the versions I've seen, which were mostly pretty good. In general, of course, I support this. -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

I myself is totally all for this imagine this plugin system in the major clients that would give Direct Connect the edge it needs since mods get unmaintained quite often this would actually help development since those mod developers could become plugin developers. -- Plugins support https://bugs

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

Here is the source for the C sample plugin. (includes visual studio project files, code::blocks project files (cross-platform ide, capable of woking with various compilers including mingw), and also windows batch file for mingw. The rest of the plugin source codes and compiled binaries can be foun

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Changed in: dcplusplus Importance: Undecided => Wishlist -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC++: New Bug description: Ok, I'll leave out the sales

[Linuxdcpp-team] [Bug 587597] Re: Plugins support

** Patch added: "against rev 2155" http://launchpadlibrarian.net/49398711/plugins.patch -- Plugins support https://bugs.launchpad.net/bugs/587597 You received this bug notification because you are a member of Dcplusplus-team, which is subscribed to DC++. Status in DC++: New Bug description: