Since we can put small icon textures in the fontstrings, perhaps we should add a media type for them.
im going to support the {star} {diamond} icon shorthands in Prat, it would be nice if I could use the SML names for the icons, ie {sml-someicon}.
I guess there will have to be some other library to break apart icons which are all in one texture (like raid icons), since you need to store the TexCoords, or the size/indexes of each icon.
There's another kind of texture (or pair of textures, actually) I think would be useful in SharedMedia. That would be textures for minimap masks and minimap borders. This would allow these kinds of textures to be shared among various addons that manipulate the minimap. The minimap mask texture would also benefit from additional metadata describing the shape of the mask, for addons which use it. (If it's impractical to add these kinds of metadata, perhaps texture contributors could agree on a naming convention that implies the mask's shape.)
Edit: I'm rethinking the utility of this now, since there can only ever be one of each texture active at any time. However, it would still facilitate contributions of new textures for minimap addons which supported them.
For the minimap textures it doesn't make sense at all to have them in SharedMedia since you tend to have only one addon running that changes shape/textures of the minimap.
wrt the icon textures: for local use these {} shortcuts could make sense but since you mention Prat I'd find it annoying if I get tons of such tags in my chat without seeing them. But simply dropping them if the icon isn't known could result in the other person expecting them... Would at least need an alt 'tag'... On the other hand I also don't want icon smilies in chat.... that will make chat even worse than it's now... :/
And as you said it will result in some problems with multiicon textures (although I think LSM is capable of storing tables, too)
Its ineveitable that they will be included in the chat.
Just provide either a 1x1 texture by default maybe, the user could actually choose whatever they want for the placeholder that way and all applications would use the same setting.
My plan for Prat is to support {SML-*} and {BLZ-*} icons, i still havent decided about multicon texturers though. There are about 5 paramters needed in order to find the right one, and those have to be given manually for the most part.
well, there are two ways:
- {SML-bla:ux:uy:vx:vy} where SML("icon", "bla") = [[path/to/texture.blp]]
- {SML-bla} where SML("icon", "bla" = {[[path//to/texture.blp]], ux, uy, vx, vy}
theoreticaly the 2nd one wouldn't be a problem since it would be the official format for this official type.
I already see people whispering me with stupid tags that i cant read because i dont have that specific addon .. oh my, please not
You have 2 choices - try to get a standard going - or do nothing and end up with a complicated addon to block all the variants. (Actually you really should just blame bliz for printing the {} that it doesnt understand)
Elkano - I like the second one because it keeps the logic at the client side, you cant trust that the numbers in the first one will be correct. Besides, I would like them to actually be able to change the icons associated with a name on thier end, so i'd like to hide any additional details which may apply to the actual texture that the name points to.
well, SML isn't really ment for dynamic asignments of data to keys but rather static data. So in the end it would propably be better to keep that stuff in Prat itself or sth like that...
well, SML isn't really ment for dynamic asignments of data to keys but rather static data. So in the end it would propably be better to keep that stuff in Prat itself or sth like that...
What stuff? remapping say new_icon1 to new_icon3? It was just a thought - Prat will just give you the ability to turn the icons off entirely, and it will show the "placeholder" for missing icons if SML ends up having one - if not it will show nothing probably.
I think at some point someone will make SharedMediaOptions which will let people remap neme->media mappings, so they apply everywhere. I could be wrong. A good example would be "Themes", which change groups of media - lets say it makes all your sounds into holiday sounds.
The easiest way is to change the meaning of they SML keys, so if you were to disable the Theming addon, your SML binding would still work. Anyhow I got on a tangent, but its not a bad idea. Would you want to work with me to create a theming system? I'm thinking of a way to provide the functionality to SML/LSM istead of trying to hook its functions or something like that. Instead, just an API I could call to register a theme "provider" which would just translate logical names from the original to its themed replacement. Unless of course you wanted to make THEME a media type (a string key mapped to a table containing key/value replacements - ie. global overrrides for each media key name)
So maybe I'm missing something (Probably not, but I'll say this for the fun of it)
But the point of embedding icons into text, or sending it between users is because you want to send something static like a spell icon, it's not like fonts or textures where it can actually be useful to let users expand their list beyond what you have.
Theres no reason to make SML support this, if your addon wants support for an icon then add a map with what the default things Blizzard requires, that way you can add something simple like say {foobar} without wasting half of the message on metadata that people without the addon are going to see.
For example, Afflicted will send {spellID} with messages, all Afflicted users have a :AddMessage hook that will call GetSpellInfo on any numbers within {} it finds, if results are returned it replaces it with the texture meta data. It's simple, Afflicted users get spell icons, non Afflicted users get a little bit of meta data that doesn't really hurt (And you can disable sending the icons anyway).
It's not like it's complicated to add your addon to the tag list, all you do is.
So, what you are saying is that, there is no use for adding icons to sml, because you can just add them to the game's list of icons.
So, for example chat smiliey's, I should just add them to the blizzard table, and not bother with SML at all, because there is no use for SML to know about them, and that if you want to share icons for any purpose at all you should use the ICON_TAG_LIST table.
Was there something in my example {SML-iconname} {BLZ-inconname} that made you think i wanted to add the blizzard icons to SML?
Its great that your solution is simple for just your addon. I hope everyone makes thier own equally different and simple solution. That will certainly lead to greater overall simplicity. Maybe someone else will put just a number in the brackets too with some completely diffrerent meaning, then you can both share that number! Even more simple!
You know when you only consider your own code everything is so much easier.
I already see people whispering me with stupid tags that i cant read because i dont have that specific addon .. oh my, please not
If you're embedding icons it should be simple, and not getting in the way. You're forgetting that everyone sees the meta data, if you're embedding {SML-IconName} thats really not convent to the people who don't have those icon names or the latest SML, on top of the fact that you then hinder the ability to read text instead of improve it.
Your argument is pointless for emotes in Prat, if you want to turn :) into a graphical emote you either need to hook something like SendChatMessage and turn :) into {SML-Smilie} (Which would be so incredibly stupid), or you're hooking something on the end clients side so if it sees :) it just turns it into the graphical emote and bypasses any sort of {SML-*} system you'd come up with, or you're going to add an entry for :) into the default Blizzard stuff so {:)} is automatically parsed and turned into the emote.
If you're seriously going to add {SML-Smilies} for every emote someone uses, even as an option when only Prat users would see the end result, that'd be very stupid given everyone sees it not just Prat users.
Frankly - I dont care if you like or dislike what I do, but spare me the insults. "...even for you".
The smiley was just an example, and i dont feel like providing any more for you to find fault with and/or pick apart, since that is what you appear to be bent on doing.
The point is do we add new ICONS to SML so everyone can use them or not.
Frankly - I dont care if you like or dislike what I do, but spare me the insults. "...even for you".
The smiley was just an example, and i dont feel like providing any more for you to find fault with and/or pick apart, since that is what you appear to be bent on doing.
The point is do we add new ICONS to SML so everyone can use them or not.
So you want to provide icons for SML, but you don't want to provide an actual example as to why they're useful? If you're adding something to a library, an actual reason beyond "Lets do it" is usually helpful.
So you want to provide icons for SML, but you don't want to provide an actual example as to why they're useful? You don't add something to a library just because, you should actually have a good reason for doing it.
...whatever you say officer
[18 09:53] <sylvanaar> i am asking elkano if he will add it there - but the "bloat" police will be watchin
Yes, I'm sorry I don't subscribe to your view of lets add everything because if at least one person needs it, it's useful.
This is a minor change to this library, there are already icon packs specific to certain mods, eg. the old cartographer icon packs. There is really no reason icons shouldnt be added, thats the whole point of this library. Otherwise i have to make SharedIconLibrary-1.0 or something.
You frankly get your viewpoint on me from the malcontents, and it goes right back to Tekkub, and his little "borg" mod crap - which seemed more due to his fallout with Ace2 than anything else.
I guess I don't think its cool to berate other authors about thier work - but I guess some people think it makes them sound smart/cool when they do it.
You know I have used your SSPVP for a long time - you could say that its "bloated" and full of crap I dont use or need, but I'm glad to still be using it, even though I only use 10% of it - which covers all my needs with just one addon.
Honestly all you have done is derail this thread. You win.
this is a good reason to redifine what we use SharedMeida for.
The name Implies that a given addon will make available it's own common graphics / sounds / textures ect to other addons that a user could use elsewhere to make a consistant UI.
So adding Icon support to LSM is a valid and quite relavent point, but it would require a slight restructure of it.
LSM shouldn't have with it any actual media of any sort as it is a libary, it should however provide access to already avaiable ingame material as it does now. It NEEDS have the ability to add ANY type of media to it's repository, be it a texture / status bar, sound, Icon, Picture, ect; The Point is valid.
@Shadowed, your quest to barate another should be overlooked as an angry person out on a witch hunt, the specific use of said media in LSM is not in question. Use of it in Prat is quite irrelavent as it was simply used as an example. One could quite easily, as i understand, use it in a Ace3-OptionsTable as well as any custom UI that is designed or made.
this is a good reason to redifine what we use SharedMeida for.
The name Implies that a given addon will make available it's own common graphics / sounds / textures ect to other addons that a user could use elsewhere to make a consistant UI.
So adding Icon support to LSM is a valid and quite relavent point, but it would require a slight restructure of it.
LSM shouldn't have with it any actual media of any sort as it is a libary, it should however provide access to already avaiable ingame material as it does now. It NEEDS have the ability to add ANY type of media to it's repository, be it a texture / status bar, sound, Icon, Picture, ect; The Point is valid.
@Shadowed, your quest to barate another should be overlooked as an angry person out on a witch hunt, the specific use of said media in LSM is not in question. Use of it in Prat is quite irrelavent as it was simply used as an example. One could quite easily, as i understand, use it in a Ace3-OptionsTable as well as any custom UI that is designed or made.
It's far more relevant and consistent to redefine SML and let people register custom types beyond background, border, font, statusbar and sound, saying you want support for something besides {SML-Name} or icons specifically is irrelevant and entirely different from what was asked in the original post, nothing wrong with making more types available for authors in the SharedMedialibrary
You're argument about berating doesn't even make sense, the specific request in the post is for something like {SML-Name}, not new types.
OrionShock, LSM already allows you to register any new type of data so the discussion is not to change LSM to allow that but rather if we want to add icon as an official datatype with a convention what to store in there.
In that case one has also to think about what we consider an icon. The textures used for spells and icons where each icon is a seperate file? Or also those icons packed up togeteher in one file like the raid target siybols. These require additional information to extract them.
WRT themes und such: currently LSM only supports a key -> data relation which is considered static and different addons should use the same key for the same file in case they both provide it (eg for fonts it's the name set in the font file). Theme support would be aliasing put ontop of that system so theoreticaly that should be possible though we'll have to look into how the API would have to change.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
im going to support the {star} {diamond} icon shorthands in Prat, it would be nice if I could use the SML names for the icons, ie {sml-someicon}.
I guess there will have to be some other library to break apart icons which are all in one texture (like raid icons), since you need to store the TexCoords, or the size/indexes of each icon.
Edit: I'm rethinking the utility of this now, since there can only ever be one of each texture active at any time. However, it would still facilitate contributions of new textures for minimap addons which supported them.
wrt the icon textures: for local use these {} shortcuts could make sense but since you mention Prat I'd find it annoying if I get tons of such tags in my chat without seeing them. But simply dropping them if the icon isn't known could result in the other person expecting them... Would at least need an alt 'tag'... On the other hand I also don't want icon smilies in chat.... that will make chat even worse than it's now... :/
And as you said it will result in some problems with multiicon textures (although I think LSM is capable of storing tables, too)
Its ineveitable that they will be included in the chat.
Just provide either a 1x1 texture by default maybe, the user could actually choose whatever they want for the placeholder that way and all applications would use the same setting.
My plan for Prat is to support {SML-*} and {BLZ-*} icons, i still havent decided about multicon texturers though. There are about 5 paramters needed in order to find the right one, and those have to be given manually for the most part.
- {SML-bla:ux:uy:vx:vy} where SML("icon", "bla") = [[path/to/texture.blp]]
- {SML-bla} where SML("icon", "bla" = {[[path//to/texture.blp]], ux, uy, vx, vy}
theoreticaly the 2nd one wouldn't be a problem since it would be the official format for this official type.
You have 2 choices - try to get a standard going - or do nothing and end up with a complicated addon to block all the variants. (Actually you really should just blame bliz for printing the {} that it doesnt understand)
Elkano - I like the second one because it keeps the logic at the client side, you cant trust that the numbers in the first one will be correct. Besides, I would like them to actually be able to change the icons associated with a name on thier end, so i'd like to hide any additional details which may apply to the actual texture that the name points to.
What stuff? remapping say new_icon1 to new_icon3? It was just a thought - Prat will just give you the ability to turn the icons off entirely, and it will show the "placeholder" for missing icons if SML ends up having one - if not it will show nothing probably.
I think at some point someone will make SharedMediaOptions which will let people remap neme->media mappings, so they apply everywhere. I could be wrong. A good example would be "Themes", which change groups of media - lets say it makes all your sounds into holiday sounds.
The easiest way is to change the meaning of they SML keys, so if you were to disable the Theming addon, your SML binding would still work. Anyhow I got on a tangent, but its not a bad idea. Would you want to work with me to create a theming system? I'm thinking of a way to provide the functionality to SML/LSM istead of trying to hook its functions or something like that. Instead, just an API I could call to register a theme "provider" which would just translate logical names from the original to its themed replacement. Unless of course you wanted to make THEME a media type (a string key mapped to a table containing key/value replacements - ie. global overrrides for each media key name)
But the point of embedding icons into text, or sending it between users is because you want to send something static like a spell icon, it's not like fonts or textures where it can actually be useful to let users expand their list beyond what you have.
Theres no reason to make SML support this, if your addon wants support for an icon then add a map with what the default things Blizzard requires, that way you can add something simple like say {foobar} without wasting half of the message on metadata that people without the addon are going to see.
For example, Afflicted will send {spellID} with messages, all Afflicted users have a :AddMessage hook that will call GetSpellInfo on any numbers within {} it finds, if results are returned it replaces it with the texture meta data. It's simple, Afflicted users get spell icons, non Afflicted users get a little bit of meta data that doesn't really hurt (And you can disable sending the icons anyway).
It's not like it's complicated to add your addon to the tag list, all you do is.
table.insert(ICON_LIST, "|TInterface\\TargetingFrame\\UI-RaidTargetingIcon_1:")
ICON_TAG_LIST["foobar"] = #(ICON_LIST)
And there you go, you now can use {foobar} to show the raid icon #1.
So, what you are saying is that, there is no use for adding icons to sml, because you can just add them to the game's list of icons.
So, for example chat smiliey's, I should just add them to the blizzard table, and not bother with SML at all, because there is no use for SML to know about them, and that if you want to share icons for any purpose at all you should use the ICON_TAG_LIST table.
Was there something in my example {SML-iconname} {BLZ-inconname} that made you think i wanted to add the blizzard icons to SML?
Its great that your solution is simple for just your addon. I hope everyone makes thier own equally different and simple solution. That will certainly lead to greater overall simplicity. Maybe someone else will put just a number in the brackets too with some completely diffrerent meaning, then you can both share that number! Even more simple!
You know when you only consider your own code everything is so much easier.
If you're embedding icons it should be simple, and not getting in the way. You're forgetting that everyone sees the meta data, if you're embedding {SML-IconName} thats really not convent to the people who don't have those icon names or the latest SML, on top of the fact that you then hinder the ability to read text instead of improve it.
Your argument is pointless for emotes in Prat, if you want to turn :) into a graphical emote you either need to hook something like SendChatMessage and turn :) into {SML-Smilie} (Which would be so incredibly stupid), or you're hooking something on the end clients side so if it sees :) it just turns it into the graphical emote and bypasses any sort of {SML-*} system you'd come up with, or you're going to add an entry for :) into the default Blizzard stuff so {:)} is automatically parsed and turned into the emote.
If you're seriously going to add {SML-Smilies} for every emote someone uses, even as an option when only Prat users would see the end result, that'd be very stupid given everyone sees it not just Prat users.
The smiley was just an example, and i dont feel like providing any more for you to find fault with and/or pick apart, since that is what you appear to be bent on doing.
The point is do we add new ICONS to SML so everyone can use them or not.
So you want to provide icons for SML, but you don't want to provide an actual example as to why they're useful? If you're adding something to a library, an actual reason beyond "Lets do it" is usually helpful.
...whatever you say officer
This is a minor change to this library, there are already icon packs specific to certain mods, eg. the old cartographer icon packs. There is really no reason icons shouldnt be added, thats the whole point of this library. Otherwise i have to make SharedIconLibrary-1.0 or something.
You frankly get your viewpoint on me from the malcontents, and it goes right back to Tekkub, and his little "borg" mod crap - which seemed more due to his fallout with Ace2 than anything else.
I guess I don't think its cool to berate other authors about thier work - but I guess some people think it makes them sound smart/cool when they do it.
You know I have used your SSPVP for a long time - you could say that its "bloated" and full of crap I dont use or need, but I'm glad to still be using it, even though I only use 10% of it - which covers all my needs with just one addon.
Honestly all you have done is derail this thread. You win.
this is a good reason to redifine what we use SharedMeida for.
The name Implies that a given addon will make available it's own common graphics / sounds / textures ect to other addons that a user could use elsewhere to make a consistant UI.
So adding Icon support to LSM is a valid and quite relavent point, but it would require a slight restructure of it.
LSM shouldn't have with it any actual media of any sort as it is a libary, it should however provide access to already avaiable ingame material as it does now. It NEEDS have the ability to add ANY type of media to it's repository, be it a texture / status bar, sound, Icon, Picture, ect; The Point is valid.
@Shadowed, your quest to barate another should be overlooked as an angry person out on a witch hunt, the specific use of said media in LSM is not in question. Use of it in Prat is quite irrelavent as it was simply used as an example. One could quite easily, as i understand, use it in a Ace3-OptionsTable as well as any custom UI that is designed or made.
It's far more relevant and consistent to redefine SML and let people register custom types beyond background, border, font, statusbar and sound, saying you want support for something besides {SML-Name} or icons specifically is irrelevant and entirely different from what was asked in the original post, nothing wrong with making more types available for authors in the SharedMedia library
You're argument about berating doesn't even make sense, the specific request in the post is for something like {SML-Name}, not new types.
In that case one has also to think about what we consider an icon. The textures used for spells and icons where each icon is a seperate file? Or also those icons packed up togeteher in one file like the raid target siybols. These require additional information to extract them.
WRT themes und such: currently LSM only supports a key -> data relation which is considered static and different addons should use the same key for the same file in case they both provide it (eg for fonts it's the name set in the font file). Theme support would be aliasing put ontop of that system so theoreticaly that should be possible though we'll have to look into how the API would have to change.