feat: add setting for showing pronouns in user info popup #5442

franzitrone · 2024-06-06T14:57:08Z

Common pronoun providers like alejo.io and pronoundb.org have respective browser extensions for twitch, but so far Chatterino does not support these providers.

I have now implemented a provider that fetches pronouns from APIs.

Pronoun providers can be configured in code, where, if multiple are configured the latter ones serve as a fallback for the former ones.

Right now, pronoundb.org and pronouns.alejo.io are implemented.

Pronouns are fetched when a message from a new user arrives and is cached application wide per session (as opposed to cache per tab). Pronouns are not added to the first message from a user. In the future, pronouns could be added to previous messages too.

For future-proofing, it is possible to fetch pronouns for multiple users at once, but right now, pronouns are fetched per-message if required.

pajlada · 2024-06-06T15:22:43Z

Some comments before I make a proper review:
Have you reached out to the API owners regarding adding this feature?
Do those API owners have some sort of privacy policy regarding data they store, and for the requests we make from Chatterino?
I'm not super fond of a request being spun off as soon as a message comes in, would it be super bad if it instead runs on a timer & bulk fetches data every ~30 seconds or so?

github-actions

clang-tidy made some suggestions

There were too many comments to post at once. Showing the first 25 out of 56. Check the log or trigger a new build to see more.

src/providers/pronouns/IPronounsApi.hpp

src/providers/pronouns/PronounUser.cpp

src/providers/pronouns/UserPronouns.hpp

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

franzitrone · 2024-06-06T17:11:14Z

Some comments before I make a proper review: Have you reached out to the API owners regarding adding this feature? Do those API owners have some sort of privacy policy regarding data they store, and for the requests we make from Chatterino? I'm not super fond of a request being spun off as soon as a message comes in, would it be super bad if it instead runs on a timer & bulk fetches data every ~30 seconds or so?

I think it should be easily possible to fetch on a timer, although I'm not familiar enough with the code base to know how or where timers are usually registered and run. Could you give me a pointer to where timers are usually placed? There was a "recent chatters" variable somewhere iirc, we could use this to occasionally on timer fetch pronouns for those.

Alejo.io unfortunately, to my understanding, has no documentation for the API at all, PronounDB has a documentation at https://pronoundb.org/wiki/api-docs and mentions setting a User-Agent (which should be set, as NetworkRequest is used). It does also provide a contact email, I'm probably going to ask there on what's the preferred way.

clang-tidy made some suggestions

There were too many comments to post at once. Showing the first 25 out of 56. Check the log or trigger a new build to see more.

I just realized I read and then completely skipped the entire clang-tidy part in the contribution guide to "do it later"... Whoops, I shall run it again locally and make the changes.

It also seems some things are broken on windows build (I think it's some std::size_t thing). I'll see if I can fix that, although I don't have Windows on any machine here.

pajlada · 2024-06-06T17:40:22Z

Any sort of nitpicks or windows/macos/linux-specific things you can't fix can be easily fixed by any of our other contributors as part of the review process - we don't 100% listen to clang-tidy, so you can make a judgement call whether you care to listen to it

For the timer thing, I can bring up some example when I do my review of this PR in the weekend.
Generally, my opinion of the feature is that it should be opt-in as a user of Chatterino to use the API, and generally we shouldn't modify old messages unless very necessary.

franzitrone · 2024-06-06T17:45:01Z

Any sort of nitpicks or windows/macos/linux-specific things you can't fix can be easily fixed by any of our other contributors as part of the review process - we don't 100% listen to clang-tidy, so you can make a judgement call whether you care to listen to it

Alright that sounds good! I'll go through the code again later or tomorrow.

For the timer thing, I can bring up some example when I do my review of this PR in the weekend.

Sounds good! When will you be reviewing? I might force-push new changes to my branch, but if needed I can just create a new branch and work on that.

Generally, my opinion of the feature is that it should be opt-in as a user of Chatterino to use the API, and generally we shouldn't modify old messages unless very necessary.

I've already created a checkbox in the settings menu to opt-out, I'll default it to disable then, which should also disable fetching.

pajlada · 2024-06-06T17:49:32Z

I should be able to get a full review in on Saturday on Sunday on my stream, can delay to Sunday (or to whenever) if you want more time before I take a look

We Squash & Merge here, so you don't have to keep commits clean, I just prefer no force pushes after I've started reviewing

Nerixyz · 2024-06-06T20:50:03Z

To fix compilation on Windows, you need to add an l, because long and unsigned long are four bytes (reference):

diff --git a/src/providers/pronouns/pronoundb/PronounsPronounDbApi.cpp b/src/providers/pronouns/pronoundb/PronounsPronounDbApi.cpp
index ca39f70f..728aed67 100644
--- a/src/providers/pronouns/pronoundb/PronounsPronounDbApi.cpp
+++ b/src/providers/pronouns/pronoundb/PronounsPronounDbApi.cpp
@@ -111,7 +111,7 @@ void PronounsPronounDbApi::fetch(std::vector<PronounUser> allUsers, IPronounsApi
   auto request = std::make_shared<IPronounsApi::RequestT>(allUsers.size(), std::move(onDone));
 
   for (std::size_t i {0}; i < allUsers.size(); i += 50) {
-    auto end = static_cast<std::size_t>(std::min(allUsers.size() - i, 50ul));
+    auto end = std::min(allUsers.size() - i, 50ull);
     auto users = std::vector<PronounUser>(allUsers.begin() + i, allUsers.begin() + i + end);
 
     NetworkRequest(PronounsPronounDbApi::makeRequest(users))

Just some general comments:

Please use clang-format to format your code (tip: turn on format-on-save in your editor).
The posted clang-tidy suggestions are valid (maybe the enum size one is debatable). You can use clang-tidy through clangd in your editor.
For documentation comments, use triple slash comments (/// - technically you could also use /** */, but it wastes two lines per comment).
Use QString over std::string, because we use it everywhere, and it's reference counted (but it doesn't have SSO sadly)
Although we mostly use one top-level namespace, I feel like a separate namespace for pronouns could be nice (so something like chatterino::PronounsPronounDbApi could become chatterino::pronouns::DbApi)
Currently, Chatterino makes one request to the DB per user, which could spam the provider quite a bit in fast chats. Maybe we could batch some requests (maybe at most one request per two seconds?).

pajlada · 2024-06-06T21:08:02Z

For documentation comments, use triple slash comments (/// - technically you could also use /** */, but it wastes two lines per comment).

Although we mostly use one top-level namespace, I feel like a separate namespace for pronouns could be nice (so something like chatterino::PronounsPronounDbApi could become chatterino::pronouns::DbApi)

You can (and probably should) leave these changes until the implementation has been reviewed

Currently, Chatterino makes one request to the DB per user, which could spam the provider quite a bit in fast chats. Maybe we could batch some requests (maybe at most one request per two seconds?).

this is the same as i mentioned above

franzitrone · 2024-06-06T21:30:02Z

I just noticed both of you had commented since I last checked (and before I pushed)!

@Nerixyz will do!

The latest version now queues users for fetching when messages arrive and are regularly dequeued and fetched in a batch. I don't think the alejo.io API supports batch requests and the (open-source) browser extension fetches per message too, but I still added a limit to avoid completely flooding it.

github-actions

clang-tidy made some suggestions

There were too many comments to post at once. Showing the first 25 out of 55. Check the log or trigger a new build to see more.

github-actions · 2024-06-06T21:30:24Z

src/providers/pronouns/IPronounsApi.hpp

+
+namespace chatterino {
+
+class IPronounsApi


warning: destructor of 'IPronounsApi' is public and non-virtual [cppcoreguidelines-virtual-class-destructor]

class IPronounsApi ^

Additional context

src/providers/pronouns/IPronounsApi.hpp:13: make it public and virtual

class IPronounsApi ^

github-actions · 2024-06-06T21:30:24Z

src/providers/pronouns/IPronounsApi.hpp

+    using RequestT = PronounsApiRequest<ResultT>;
+
+    // Fetch pronouns from the API.
+    virtual void fetch(std::vector<PronounUser> users,


warning: the parameter 'users' is copied for each invocation but only used as a const reference; consider making it a const reference [performance-unnecessary-value-param]

virtual void fetch(std::vector<PronounUser> users, ^

github-actions · 2024-06-06T21:30:24Z

src/providers/pronouns/IPronounsApi.hpp

+
+    // Fetch pronouns from the API.
+    virtual void fetch(std::vector<PronounUser> users,
+                       RequestT::CallbackT &&onDone){};


warning: rvalue reference parameter 'onDone' is never moved from inside the function body [cppcoreguidelines-rvalue-reference-param-not-moved]

RequestT::CallbackT &&onDone){}; ^

github-actions · 2024-06-06T21:30:24Z

src/providers/pronouns/PronounUser.cpp

+namespace chatterino {
+
+PronounUser::PronounUser()
+    : id{}


warning: initializer for member 'id' is redundant [readability-redundant-member-init]

Suggested change

: id{}

:

github-actions · 2024-06-06T21:30:24Z

src/providers/pronouns/PronounUser.cpp

+
+PronounUser::PronounUser()
+    : id{}
+    , username{}


warning: initializer for member 'username' is redundant [readability-redundant-member-init]

Suggested change

, username{}

,

github-actions · 2024-06-06T21:30:26Z

src/providers/pronouns/UserPronouns.cpp

+}
+UserPronouns::UserPronouns(std::string pronouns)
+{
+    if (pronouns.length() == 0)


warning: the 'empty' method should be used to check for emptiness instead of 'length' [readability-container-size-empty]

Suggested change

if (pronouns.length() == 0)

if (pronouns.empty())

Additional context

/usr/include/c++/13/bits/basic_string.h:1219: method 'basic_string'::empty() defined here

empty() const _GLIBCXX_NOEXCEPT ^

github-actions · 2024-06-06T21:30:26Z

src/providers/pronouns/UserPronouns.hpp

+    UserPronouns();
+    UserPronouns(std::string);
+
+    inline std::string format() const


warning: function 'format' has inline specifier but is implicitly inlined [readability-redundant-inline-specifier]

Suggested change

inline std::string format() const

std::string format() const

src/providers/pronouns/UserPronouns.hpp

github-actions · 2024-06-06T21:30:26Z

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

+/*static*/ std::optional<std::unordered_map<std::string, std::string>>
+    PronounsAlejoApi::pronounsFromId = {};
+
+/*static*/ UserPronouns PronounsAlejoApi::parse(QJsonObject object)


warning: method 'parse' can be made static [readability-convert-member-functions-to-static]

Suggested change

/*static*/ UserPronouns PronounsAlejoApi::parse(QJsonObject object)

/*static*/ static UserPronouns PronounsAlejoApi::parse(QJsonObject object)

github-actions · 2024-06-06T21:30:26Z

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

+/*static*/ std::optional<std::unordered_map<std::string, std::string>>
+    PronounsAlejoApi::pronounsFromId = {};
+
+/*static*/ UserPronouns PronounsAlejoApi::parse(QJsonObject object)


warning: out-of-line definition of 'parse' does not match any declaration in 'chatterino::PronounsAlejoApi' [clang-diagnostic-error]

/*static*/ UserPronouns PronounsAlejoApi::parse(QJsonObject object) ^

Felanbird · 2024-06-06T21:35:21Z

I don't think the alejo.io API supports batch requests and the (open-source) browser extension fetches per message too, but I still added a limit to avoid completely flooding it.

My main issue I outlined in #3346 was that alejo almost certainly can't keep up with the amount of requests we'd send, that 2 different implementations would return a different subset of users with pronouns, and it falls to blaming Chatterino if this feature we add works very sporadically (e.g. user reaction to bttv live emote updates)

pronounsdb does look like it could better handle the brunt of our requests in comparison though

franzitrone · 2024-06-06T21:45:39Z

I don't think the alejo.io API supports batch requests and the (open-source) browser extension fetches per message too, but I still added a limit to avoid completely flooding it.

My main issue I outlined in #3346 was that alejo almost certainly can't keep up with the amount of requests we'd send, that 2 different implementations would return a different subset of users with pronouns, and it falls to blaming Chatterino if this feature we add works very sporadically (e.g. user reaction to bttv live emote updates)

pronounsdb does look like it could better handle the brunt of our requests in comparison though

PronounDB really does seem to handle it better, but it is unfortunately less widely used at least in the channels I've been lurking in. I'm wondering how alejo.io is working in the first place, as it seems that the browser extension is working in a flooding way too. If this is the intended way, and the way it is already implemented in its own browser extension, I don't see what sets apart Chatterino from just using a browser with the extension enabled (and then flooding the server in very lively channels).

Felanbird · 2024-06-06T21:47:02Z

I don't see what sets apart Chatterino from just using a browser with the extension enabled

Browser chat is 1 channel and Chatterino is up to 100

franzitrone · 2024-06-06T21:50:16Z

I don't see what sets apart Chatterino from just using a browser with the extension enabled

Browser chat is 1 channel and Chatterino is up to 100

True. I suppose only queuing new users on message in the active tab leads to the aforementioned weird sporadic behaviour?

iProdigy · 2024-06-06T22:32:14Z

My recommendation is to only display the pronouns in the usercard - this minimizes http requests and avoids implementation complexity (of batching requests or caching thousands of pronouns)

I understand that the usercard approach reduces the convenience factor for this feature, but the technical challenges (in my opinion) puts us in a situation where we need to apply more pressure on Twitch to natively support pronouns (to display it inline with each chat message). Similar to the no-video and no-audio badges, there should be he,she,they badges natively selectable on Twitch, which would be very easy for us to support in a scalable manner

Felanbird · 2024-06-06T23:08:34Z

there should be he,she,they badges natively selectable on Twitch, which would be very easy for us to support in a scalable manner

not even just a badge, pronouns sent in the tags would be fine too (don't care about people who hate tags, they work)

github-actions

clang-tidy made some suggestions

There were too many comments to post at once. Showing the first 25 out of 53. Check the log or trigger a new build to see more.

github-actions · 2024-06-07T12:04:20Z

src/providers/pronouns/IPronounsApi.hpp

+
+namespace chatterino {
+
+class IPronounsApi


warning: destructor of 'IPronounsApi' is public and non-virtual [cppcoreguidelines-virtual-class-destructor]

class IPronounsApi ^

Additional context

src/providers/pronouns/IPronounsApi.hpp:13: make it public and virtual

class IPronounsApi ^

github-actions · 2024-06-07T12:04:21Z

src/providers/pronouns/IPronounsApi.hpp

+    using RequestT = PronounsApiRequest<ResultT>;
+
+    // Fetch pronouns from the API.
+    virtual void fetch(std::vector<PronounUser> users,


warning: the parameter 'users' is copied for each invocation but only used as a const reference; consider making it a const reference [performance-unnecessary-value-param]

virtual void fetch(std::vector<PronounUser> users, ^

github-actions · 2024-06-07T12:04:21Z

src/providers/pronouns/IPronounsApi.hpp

+
+    // Fetch pronouns from the API.
+    virtual void fetch(std::vector<PronounUser> users,
+                       RequestT::CallbackT &&onDone){};


warning: rvalue reference parameter 'onDone' is never moved from inside the function body [cppcoreguidelines-rvalue-reference-param-not-moved]

RequestT::CallbackT &&onDone){}; ^

github-actions · 2024-06-07T12:04:21Z

src/providers/pronouns/PronounUser.cpp

+namespace chatterino {
+
+PronounUser::PronounUser()
+    : id{}


warning: initializer for member 'id' is redundant [readability-redundant-member-init]

Suggested change

: id{}

:

github-actions · 2024-06-07T12:04:21Z

src/providers/pronouns/PronounUser.cpp

+
+PronounUser::PronounUser()
+    : id{}
+    , username{}


warning: initializer for member 'username' is redundant [readability-redundant-member-init]

Suggested change

, username{}

,

github-actions · 2024-06-07T12:04:23Z

src/providers/pronouns/UserPronouns.hpp

+    UserPronouns();
+    UserPronouns(std::string);
+
+    inline std::string format() const


warning: function 'format' has inline specifier but is implicitly inlined [readability-redundant-inline-specifier]

Suggested change

inline std::string format() const

std::string format() const

src/providers/pronouns/UserPronouns.hpp

github-actions · 2024-06-07T12:04:23Z

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

+/*static*/ std::optional<std::unordered_map<std::string, std::string>>
+    PronounsAlejoApi::pronounsFromId = {};
+
+/*static*/ UserPronouns PronounsAlejoApi::parse(QJsonObject object)


warning: method 'parse' can be made static [readability-convert-member-functions-to-static]

Suggested change

/*static*/ UserPronouns PronounsAlejoApi::parse(QJsonObject object)

/*static*/ static UserPronouns PronounsAlejoApi::parse(QJsonObject object)

github-actions · 2024-06-07T12:04:23Z

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

+    else
+    {
+        return {};
+    }


warning: do not use 'else' after 'return' [llvm-else-after-return]

Suggested change

}

return {};

github-actions · 2024-06-07T12:04:23Z

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

+        NetworkRequest(PronounsAlejoApi::API_URL + PronounsAlejoApi::API_USERS +
+                       "/" + user.username)
+            .concurrent()
+            .onSuccess([request, user](auto result) {


warning: no viable conversion from '(lambda at /github/workspace/src/providers/pronouns/alejo/PronounsAlejoApi.cpp:134:24)' to 'NetworkSuccessCallback' (aka 'function<void (NetworkResult)>') [clang-diagnostic-error]

.onSuccess([request, user](auto result) { ^

Additional context

/usr/include/c++/13/bits/std_function.h:374: candidate constructor not viable: no known conversion from '(lambda at /github/workspace/src/providers/pronouns/alejo/PronounsAlejoApi.cpp:134:24)' to 'nullptr_t' (aka 'std::nullptr_t') for 1st argument

function(nullptr_t) noexcept ^

/usr/include/c++/13/bits/std_function.h:385: candidate constructor not viable: no known conversion from '(lambda at /github/workspace/src/providers/pronouns/alejo/PronounsAlejoApi.cpp:134:24)' to 'const function<void (NetworkResult)> &' for 1st argument

function(const function& __x) ^

/usr/include/c++/13/bits/std_function.h:403: candidate constructor not viable: no known conversion from '(lambda at /github/workspace/src/providers/pronouns/alejo/PronounsAlejoApi.cpp:134:24)' to 'function<void (NetworkResult)> &&' for 1st argument

function(function&& __x) noexcept ^

/usr/include/c++/13/bits/std_function.h:434: candidate template ignored: requirement '_Callable<(lambda at /github/workspace/src/providers/pronouns/alejo/PronounsAlejoApi.cpp:134:24), (lambda at /github/workspace/src/providers/pronouns/alejo/PronounsAlejoApi.cpp:134:24), std::__invoke_result<(lambda at /github/workspace/src/providers/pronouns/alejo/PronounsAlejoApi.cpp:134:24) &, chatterino::NetworkResult>>::value' was not satisfied [with _Functor = (lambda at /github/workspace/src/providers/pronouns/alejo/PronounsAlejoApi.cpp:134:24)]

function(_Functor&& __f) ^

src/common/network/NetworkRequest.hpp:46: passing argument to parameter 'cb' here

NetworkRequest onSuccess(NetworkSuccessCallback cb) &&; ^

franzitrone · 2024-06-08T08:26:11Z

I don't think I'll be able to update the PR this weekend, but I've installed the clang LS extension and I starting working through the suggestions yesterday.

The owner of PronounDB also replied yesterday I'll read it carefully tomorrow or so and share her comments.

Apologies for accidentally closing the PR yesterday, I was scrolling through the comments and somehow clicked "Close" 😒

pajlada · 2024-06-08T09:28:54Z

Alright, I'll wait with reviewing the PR until you've had a chance to look at it further then. Poke me!

treuks · 2024-06-12T13:43:16Z

My recommendation is to only display the pronouns in the usercard - this minimizes http requests and avoids implementation complexity (of batching requests or caching thousands of pronouns)

There are certain pronoundb plugins for Discord, which show pronouns for a user. They show the pronouns inline with the messages, but only after you click on their profile, like so:

User message before you open the profile:

User message after you open the profile:

In my opinion this is a fairly decent approach to this sort of situation, we could do it like so:

Before:

Open user card:

After:

I feel like this could be a decent middleground between only showing it in the usercard and making requests for like every user. thoughts?

pajlada · 2024-06-12T13:48:56Z

I would be fine with that compromise @treuks

iProdigy · 2024-06-12T17:34:23Z

compromise is good; i'm not a maintainer but could be easier to review if 2 separate prs

franzitrone · 2024-08-29T14:29:59Z

Awesome! I'll try to merge and remove the draft tag.

…card-alejoio

github-actions

clang-tidy made some suggestions

github-actions · 2024-08-29T15:20:40Z

benchmarks/src/RecentMessages.cpp

@@ -124,6 +129,7 @@ class MockApplication : public mock::BaseApplication
    FfzEmotes ffzEmotes;
    SeventvEmotes seventvEmotes;
    DisabledStreamerMode streamerMode;
+    Pronouns pronouns;


warning: field has incomplete type 'chatterino::pronouns::Pronouns' [clang-diagnostic-error]

Pronouns pronouns; ^

Additional context

src/Application.hpp:56: forward declaration of 'chatterino::pronouns::Pronouns'

class Pronouns; ^

github-actions · 2024-08-29T15:20:40Z

benchmarks/src/RecentMessages.cpp

@@ -124,6 +129,7 @@
    FfzEmotes ffzEmotes;
    SeventvEmotes seventvEmotes;
    DisabledStreamerMode streamerMode;
+    Pronouns pronouns;


warning: unknown type name 'Pronouns'; did you mean 'chatterino::pronouns::Pronouns'? [clang-diagnostic-error]

Suggested change

Pronouns pronouns;

chatterino::pronouns::Pronouns pronouns;

Additional context

src/Application.hpp:56: 'chatterino::pronouns::Pronouns' declared here

class Pronouns; ^

mocks/include/mocks/EmptyApplication.hpp

github-actions · 2024-08-29T15:20:41Z

tests/src/InputCompletion.cpp

    AccountController accounts;
    mock::MockTwitchIrcServer twitch;
    Emotes emotes;
    BttvEmotes bttvEmotes;
    FfzEmotes ffzEmotes;
    SeventvEmotes seventvEmotes;
+    Pronouns pronouns;


warning: field has incomplete type 'chatterino::pronouns::Pronouns' [clang-diagnostic-error]

Pronouns pronouns; ^

Additional context

src/Application.hpp:56: forward declaration of 'chatterino::pronouns::Pronouns'

class Pronouns; ^

github-actions · 2024-08-29T15:20:41Z

tests/src/InputCompletion.cpp

    AccountController accounts;
    mock::MockTwitchIrcServer twitch;
    Emotes emotes;
    BttvEmotes bttvEmotes;
    FfzEmotes ffzEmotes;
    SeventvEmotes seventvEmotes;
+    Pronouns pronouns;


warning: unknown type name 'Pronouns'; did you mean 'chatterino::pronouns::Pronouns'? [clang-diagnostic-error]

Suggested change

Pronouns pronouns;

chatterino::pronouns::Pronouns pronouns;

Additional context

src/Application.hpp:56: 'chatterino::pronouns::Pronouns' declared here

class Pronouns; ^

github-actions · 2024-08-29T15:20:42Z

tests/src/MessageBuilder.cpp

@@ -91,6 +91,20 @@ class MockApplication : public mock::BaseApplication
        return &this->seventvEmotes;
    }

+<<<<<<< HEAD:tests/src/TwitchMessageBuilder.cpp


warning: version control conflict marker in file [clang-diagnostic-error]

<<<<<<< HEAD:tests/src/TwitchMessageBuilder.cpp ^

github-actions · 2024-08-29T15:20:42Z

tests/src/MessageBuilder.cpp

@@ -102,6 +116,7 @@
    BttvEmotes bttvEmotes;
    FfzEmotes ffzEmotes;
    SeventvEmotes seventvEmotes;
+    Pronouns::Pronouns pronouns;


warning: field has incomplete type 'pronouns::Pronouns' [clang-diagnostic-error]

Pronouns::Pronouns pronouns; ^

Additional context

src/Application.hpp:56: forward declaration of 'chatterino::pronouns::Pronouns'

class Pronouns; ^

github-actions · 2024-08-29T15:20:42Z

tests/src/MessageBuilder.cpp

@@ -102,6 +116,7 @@
    BttvEmotes bttvEmotes;
    FfzEmotes ffzEmotes;
    SeventvEmotes seventvEmotes;
+    Pronouns::Pronouns pronouns;


warning: use of undeclared identifier 'Pronouns'; did you mean 'pronouns'? [clang-diagnostic-error]

Suggested change

Pronouns::Pronouns pronouns;

pronouns::Pronouns pronouns;

Additional context

src/Application.hpp:55: 'pronouns' declared here

namespace pronouns { ^

Mm2PL · 2024-08-29T15:13:34Z

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

+    qCDebug(chatterinoPronouns) << "Fetching available pronouns for alejo.io";
+    NetworkRequest(AlejoApi::API_URL + AlejoApi::API_PRONOUNS)
+        .concurrent()
+        .cache()


How well or not well does this handle caching?
From my cursory glance at the caching system, I can see a scenario where:

the data gets loaded and saved to cache

a user adds themselves to the database

this gets loaded from cache but still has the old data.

Even restarting Chatterino probably wouldn't help because the cached data will still be in cache and it will just be reloaded.

This is only for the pronouns (i.e. some-pronoun-id → "they/them") not for the users, thus it's not frequently updated. But I agree that it should be removed, because we don't have any way of invalidating this.

tests/src/MessageBuilder.cpp

franzitrone · 2024-08-29T17:05:58Z

How can I run the benchmarks, tests, etc. on my machine? I'm getting Unknown CMake command "add_sanitizers".

In any case, I've pushed some changes that should hopefully fix the build errors.

Nerixyz · 2024-08-29T19:39:39Z

I'm getting Unknown CMake command "add_sanitizers".

Are you sure you properly checked out all submodules (git submodule update --init --recursive)?

franzitrone · 2024-08-29T19:43:31Z

I'm getting Unknown CMake command "add_sanitizers".

Are you sure you properly checked out all submodules (git submodule update --init --recursive)?

I didn't, that was definitely it.

github-actions

clang-tidy made some suggestions

benchmarks/src/RecentMessages.cpp

mocks/include/mocks/EmptyApplication.hpp

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

github-actions · 2024-08-30T08:25:58Z

tests/src/InputCompletion.cpp

    AccountController accounts;
    mock::MockTwitchIrcServer twitch;
    Emotes emotes;
    BttvEmotes bttvEmotes;
    FfzEmotes ffzEmotes;
    SeventvEmotes seventvEmotes;
+    pronouns::Pronouns pronouns;


warning: field has incomplete type 'pronouns::Pronouns' [clang-diagnostic-error]

pronouns::Pronouns pronouns; ^

Additional context

src/Application.hpp:56: forward declaration of 'chatterino::pronouns::Pronouns'

class Pronouns; ^

github-actions · 2024-08-30T08:25:58Z

tests/src/MessageBuilder.cpp

@@ -102,6 +108,7 @@ class MockApplication : public mock::BaseApplication
    BttvEmotes bttvEmotes;
    FfzEmotes ffzEmotes;
    SeventvEmotes seventvEmotes;
+    pronouns::Pronouns pronouns;


warning: field has incomplete type 'pronouns::Pronouns' [clang-diagnostic-error]

pronouns::Pronouns pronouns; ^

Additional context

src/Application.hpp:56: forward declaration of 'chatterino::pronouns::Pronouns'

class Pronouns; ^

pajlada · 2024-08-31T11:46:23Z

@DeltaTimo Could you ensure this option is enabled?

If it already is, could you uncheck & check it?

franzitrone · 2024-09-01T00:54:49Z

@DeltaTimo Could you ensure this option is enabled?

If it already is, could you uncheck & check it?

Yes, sure, I'll do in a sec!

Edit: done!

pajlada · 2024-09-01T09:18:59Z

Seems like I still can't push things, weird

Here's a patch you can apply which will make it build

diff --git a/benchmarks/src/RecentMessages.cpp b/benchmarks/src/RecentMessages.cpp
index fd3461a7d..994c88858 100644
--- a/benchmarks/src/RecentMessages.cpp
+++ b/benchmarks/src/RecentMessages.cpp
@@ -11,6 +11,7 @@
 #include "providers/chatterino/ChatterinoBadges.hpp"
 #include "providers/ffz/FfzBadges.hpp"
 #include "providers/ffz/FfzEmotes.hpp"
+#include "providers/pronouns/Pronouns.hpp"
 #include "providers/recentmessages/Impl.hpp"
 #include "providers/seventv/SeventvBadges.hpp"
 #include "providers/seventv/SeventvEmotes.hpp"
diff --git a/src/Application.cpp b/src/Application.cpp
index 7f805fa1d..24dd346fd 100644
--- a/src/Application.cpp
+++ b/src/Application.cpp
@@ -187,36 +187,6 @@ Application::Application(Settings &_settings, const Paths &paths,
 
 Application::~Application()
 {
-#ifdef CHATTERINO_HAVE_PLUGINS
-    this->plugins.reset();
-#endif
-    this->twitchPubSub.reset();
-    this->twitchBadges.reset();
-    this->twitchLiveController.reset();
-    this->chatterinoBadges.reset();
-    this->bttvEmotes.reset();
-    this->ffzEmotes.reset();
-    this->seventvEmotes.reset();
-    this->notifications.reset();
-    this->commands.reset();
-    // If a crash happens after crashHandler has been reset, we'll assert
-    // This isn't super different from before, where if the app is already killed, the getApp() portion of it is already dead
-    this->crashHandler.reset();
-    this->seventvAPI.reset();
-    this->highlights.reset();
-    this->seventvBadges.reset();
-    this->ffzBadges.reset();
-    // this->twitch.reset();
-    this->imageUploader.reset();
-    this->hotkeys.reset();
-    this->fonts.reset();
-    this->sound.reset();
-    this->userData.reset();
-    this->toasts.reset();
-    this->accounts.reset();
-    this->emotes.reset();
-    this->themes.reset();
-    this->pronouns.reset();
     // we do this early to ensure getApp isn't used in any dtors
     INSTANCE = nullptr;
 }
diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
index 5ed9e2457..8aec2c5bb 100644
--- a/src/CMakeLists.txt
+++ b/src/CMakeLists.txt
@@ -348,12 +348,12 @@ set(SOURCE_FILES
         providers/liveupdates/BasicPubSubManager.hpp
         providers/liveupdates/BasicPubSubWebsocket.hpp
 
-        providers/pronouns/Pronouns.hpp
         providers/pronouns/Pronouns.cpp
-        providers/pronouns/UserPronouns.hpp
+        providers/pronouns/Pronouns.hpp
         providers/pronouns/UserPronouns.cpp
-        providers/pronouns/alejo/PronounsAlejoApi.hpp
+        providers/pronouns/UserPronouns.hpp
         providers/pronouns/alejo/PronounsAlejoApi.cpp
+        providers/pronouns/alejo/PronounsAlejoApi.hpp
 
         providers/recentmessages/Api.cpp
         providers/recentmessages/Api.hpp
diff --git a/src/providers/pronouns/alejo/PronounsAlejoApi.cpp b/src/providers/pronouns/alejo/PronounsAlejoApi.cpp
index 0d2cfe834..31e606016 100644
--- a/src/providers/pronouns/alejo/PronounsAlejoApi.cpp
+++ b/src/providers/pronouns/alejo/PronounsAlejoApi.cpp
@@ -114,7 +114,7 @@ void AlejoApi::fetch(const QString &username,
                 return;
             }
             qCWarning(chatterinoPronouns)
-                << "alejo.io returned " << status
+                << "alejo.io returned " << status.value_or(-1)
                 << " when fetching pronouns for " << username;
             onDone({});
         })
diff --git a/tests/src/InputCompletion.cpp b/tests/src/InputCompletion.cpp
index cbf4fcc65..460fb1526 100644
--- a/tests/src/InputCompletion.cpp
+++ b/tests/src/InputCompletion.cpp
@@ -69,18 +69,12 @@ public:
         return &this->seventvEmotes;
     }
 
-    pronouns::Pronouns *getPronouns() override
-    {
-        return &this->pronouns;
-    }
-
     AccountController accounts;
     mock::MockTwitchIrcServer twitch;
     Emotes emotes;
     BttvEmotes bttvEmotes;
     FfzEmotes ffzEmotes;
     SeventvEmotes seventvEmotes;
-    pronouns::Pronouns pronouns;
 };
 
 }  // namespace
diff --git a/tests/src/MessageBuilder.cpp b/tests/src/MessageBuilder.cpp
index c2a68d16a..39f01f196 100644
--- a/tests/src/MessageBuilder.cpp
+++ b/tests/src/MessageBuilder.cpp
@@ -91,12 +91,6 @@ public:
         return &this->seventvEmotes;
     }
 
-    pronouns::Pronouns *getPronouns() override
-    {
-        return &this->pronouns;
-    }
-
-    Settings settings;
     AccountController accounts;
     Emotes emotes;
     mock::UserDataController userData;
@@ -108,7 +102,6 @@ public:
     BttvEmotes bttvEmotes;
     FfzEmotes ffzEmotes;
     SeventvEmotes seventvEmotes;
-    pronouns::Pronouns pronouns;
 };
 
 }  // namespace

Felanbird · 2024-09-01T17:00:46Z

Seems like I still can't push things, weird

probably because this branch is on a chatterino7 fork, not chatterino2

franzitrone · 2024-09-01T22:16:25Z

Seems like I still can't push things, weird

probably because this branch is on a chatterino7 fork, not chatterino2

Oh yeah that's possible. I just applied the patch and pushed, thanks!

…card-alejoio

github-actions

clang-tidy made some suggestions

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

github-actions

clang-tidy made some suggestions

github-actions · 2024-09-07T12:56:16Z

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

+}
+
+void AlejoApi::fetch(const QString &username,
+                     std::function<void(std::optional<UserPronouns>)> onDone)


warning: the parameter 'onDone' is copied for each invocation but only used as a const reference; consider making it a const reference [performance-unnecessary-value-param]

src/providers/pronouns/alejo/PronounsAlejoApi.hpp:21:

- std::function<void(std::optional<UserPronouns>)> onDone); + const std::function<void(std::optional<UserPronouns>)>& onDone);

Suggested change

std::function<void(std::optional<UserPronouns>)> onDone)

const std::function<void(std::optional<UserPronouns>)>& onDone)

pajlada

Some nitpicking, will take another round of checks once these things have been done.

Functionally, the PR works almost as I'd expect. Thanks for taking the time to look into this, and I look forward to merge this in

src/Application.hpp

src/Application.cpp

src/providers/pronouns/Pronouns.cpp

pajlada · 2024-09-07T12:52:18Z

src/providers/pronouns/Pronouns.cpp

+    qCDebug(chatterinoPronouns)
+        << "Fetching pronouns from alejo.io for " << username;
+
+    alejoApi.fetch(username, [this, callbackSuccess, callbackFail,


Just confirming this is correct, and that you can ignore this. I will fix this in a follow-up PR

src/providers/pronouns/Pronouns.hpp

src/providers/pronouns/alejo/PronounsAlejoApi.hpp

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

pajlada · 2024-09-07T13:08:17Z

src/providers/pronouns/alejo/PronounsAlejoApi.cpp

+    }
+
+    qCDebug(chatterinoPronouns) << "Fetching available pronouns for alejo.io";
+    NetworkRequest(AlejoApi::API_URL + AlejoApi::API_PRONOUNS)


I would prefer if this network request is only made if the pronoun setting is enabled.
For reference, check out the BttvEmotes ctor.
If you want the request to be made in the constructor, which I think is fine for a network request this small, leave the autoInvoke param out (or set to true, its default value)

pajlada

Since this is functionally mostly good, I will merge this in and follow-up with another PR cleaning some things up

pajlada · 2024-09-08T09:20:59Z

Thank you! As a first-time contributor, you are now eligible to be listed in the About page in Chatterino.

If you want this, you can open a new PR where you modify the resources/contributors.txt file and add yourself to the list. Read the comments at the top for instructions.

pajlada self-assigned this Jun 6, 2024

github-actions bot reviewed Jun 6, 2024

View reviewed changes

franzitrone changed the title ~~feat: Fetch and show pronouns for chatters.~~ Draft: feat: Fetch and show pronouns for chatters. Jun 6, 2024

franzitrone marked this pull request as draft June 6, 2024 17:11

franzitrone changed the title ~~Draft: feat: Fetch and show pronouns for chatters.~~ feat: Fetch and show pronouns for chatters. Jun 6, 2024

franzitrone force-pushed the mr/chatterino2/add-pronoun-provider branch 2 times, most recently from 488b6a7 to d72b09c Compare June 6, 2024 21:16

github-actions bot reviewed Jun 6, 2024

View reviewed changes

franzitrone closed this Jun 7, 2024

franzitrone reopened this Jun 7, 2024

github-actions bot reviewed Jun 7, 2024

View reviewed changes

franzitrone added 2 commits August 29, 2024 16:50

Merge branch 'master' into mr/pronouns-in-usercard-alejoio

b48d594

Merge remote-tracking branch 'origin/master' into mr/pronouns-in-user…

4521572

…card-alejoio

franzitrone changed the title ~~feat: Fetch and show pronouns for chatters.~~ Fetch and show pronouns for chatters. Aug 29, 2024

franzitrone marked this pull request as ready for review August 29, 2024 15:04

github-actions bot reviewed Aug 29, 2024

View reviewed changes

Mm2PL reviewed Aug 29, 2024

View reviewed changes

franzitrone added 3 commits August 29, 2024 18:49

Fix incorrect pronouns class name.

261382a

Don't cache when fetching available pronouns from alejo.io.

4eddcf7

Remove a removed function that leaked through a merge.

612fda7

github-actions bot reviewed Aug 30, 2024

View reviewed changes

Apply patch from pajlada.

3cedf4e

Merge remote-tracking branch 'origin/master' into mr/pronouns-in-user…

fb05076

…card-alejoio

github-actions bot reviewed Sep 2, 2024

View reviewed changes

src/providers/pronouns/alejo/PronounsAlejoApi.cpp Show resolved Hide resolved

Merge branch 'master' into mr/chatterino2/add-pronoun-provider

8362afb

github-actions bot reviewed Sep 7, 2024

View reviewed changes

pajlada requested changes Sep 7, 2024

View reviewed changes

pajlada approved these changes Sep 8, 2024

View reviewed changes

pajlada changed the title ~~Fetch and show pronouns for chatters.~~ feat: add setting for showing pronouns in user info popup Sep 8, 2024

pajlada merged commit 9375bce into Chatterino:master Sep 8, 2024
18 checks passed

iProdigy mentioned this pull request Oct 15, 2024

Pronoun extension #2319

Closed

	/static/ UserPronouns PronounsAlejoApi::parse(QJsonObject object)
	/static/ static UserPronouns PronounsAlejoApi::parse(QJsonObject object)

	std::function<void(std::optional<UserPronouns>)> onDone)
	const std::function<void(std::optional<UserPronouns>)>& onDone)

feat: add setting for showing pronouns in user info popup #5442

feat: add setting for showing pronouns in user info popup #5442

Conversation

franzitrone commented Jun 6, 2024

pajlada commented Jun 6, 2024

github-actions bot left a comment

Choose a reason for hiding this comment

franzitrone commented Jun 6, 2024

pajlada commented Jun 6, 2024

franzitrone commented Jun 6, 2024

pajlada commented Jun 6, 2024

Nerixyz commented Jun 6, 2024

pajlada commented Jun 6, 2024 • edited Loading

franzitrone commented Jun 6, 2024

github-actions bot left a comment

Choose a reason for hiding this comment

github-actions bot Jun 6, 2024

Choose a reason for hiding this comment

github-actions bot Jun 6, 2024

Choose a reason for hiding this comment

github-actions bot Jun 6, 2024

Choose a reason for hiding this comment

github-actions bot Jun 6, 2024

Choose a reason for hiding this comment

github-actions bot Jun 6, 2024

Choose a reason for hiding this comment

github-actions bot Jun 6, 2024

Choose a reason for hiding this comment

github-actions bot Jun 6, 2024

Choose a reason for hiding this comment

github-actions bot Jun 6, 2024

Choose a reason for hiding this comment

github-actions bot Jun 6, 2024

Choose a reason for hiding this comment

Felanbird commented Jun 6, 2024

franzitrone commented Jun 6, 2024

Felanbird commented Jun 6, 2024

franzitrone commented Jun 6, 2024

iProdigy commented Jun 6, 2024 • edited Loading

Felanbird commented Jun 6, 2024

github-actions bot left a comment

Choose a reason for hiding this comment

github-actions bot Jun 7, 2024

Choose a reason for hiding this comment

github-actions bot Jun 7, 2024

Choose a reason for hiding this comment

github-actions bot Jun 7, 2024

Choose a reason for hiding this comment

github-actions bot Jun 7, 2024

Choose a reason for hiding this comment

github-actions bot Jun 7, 2024

Choose a reason for hiding this comment

github-actions bot Jun 7, 2024

Choose a reason for hiding this comment

github-actions bot Jun 7, 2024

Choose a reason for hiding this comment

github-actions bot Jun 7, 2024

Choose a reason for hiding this comment

github-actions bot Jun 7, 2024

Choose a reason for hiding this comment

franzitrone commented Jun 8, 2024

pajlada commented Jun 8, 2024

treuks commented Jun 12, 2024 • edited Loading

pajlada commented Jun 12, 2024

iProdigy commented Jun 12, 2024

franzitrone commented Aug 29, 2024

github-actions bot left a comment

Choose a reason for hiding this comment

github-actions bot Aug 29, 2024

Choose a reason for hiding this comment

github-actions bot Aug 29, 2024

Choose a reason for hiding this comment

github-actions bot Aug 29, 2024

Choose a reason for hiding this comment

github-actions bot Aug 29, 2024

Choose a reason for hiding this comment

github-actions bot Aug 29, 2024

Choose a reason for hiding this comment

github-actions bot Aug 29, 2024

Choose a reason for hiding this comment

pajlada commented Jun 6, 2024 •

edited

Loading

iProdigy commented Jun 6, 2024 •

edited

Loading

treuks commented Jun 12, 2024 •

edited

Loading

franzitrone commented Aug 29, 2024 •

edited

Loading

franzitrone commented Sep 1, 2024 •

edited

Loading