Fix small Wire runtime edge cases by ShiroKSH · Pull Request #3649 · wiremod/wire

ShiroKSH · 2026-07-09T15:49:46Z

splited

remaining unsure changes

blueshank-gh · 2026-07-09T16:10:58Z

I don't think this is necessary.
The != operator is provided by garry's mod, we should have a gmod-based linter already in here.
Your LLM may not have context to understand that the language is a modified version to include c-style operators.

Make sure that changes like these don't appear in PRs as its considered a "style change" which is not necessary in a feature-based PR, more so a different PR would be required for this.

This can make histories difficult to view in blame as well, I suggest reverting this.

ok thanks thats not mine fix i deleted

blueshank-gh · 2026-07-09T16:18:12Z

Also if PR's are still in progress, i.e. your still finding things to change, you should make it a draft, its there to help you before releasing to the public to review.

Grocel

Please separate "Passed" and "OK" changes into a new PR. Keep the "Unsure" and questionable changes to this PR.

Grocel · 2026-07-09T16:40:24Z

-	if not IsValid(player) then return end
-	broadcastSignal(self.data.signalgroup, name, 1, self.entity, player)
+	if not IsValid(player) or not player:IsPlayer() then return end
+	broadcastSignal(self.data.signalgroup, name, 1, self.entity, player:UniqueID())


Not passed. As far as I have been told player:UniqueID() is not expected there. This needs testing.

https://github.com/ShiroKSH/wire/blob/452f57412fc23530d3b2892d0ad2f4b1744e2a71/lua/entities/gmod_wire_expression2/core/signal.lua#L72

Looks correct

Grocel · 2026-07-09T16:40:53Z

 			for name, contexts in pairs_ac(signals) do
 				-- to remove all signals the chip registered for.
-				contexts[self] = nil
+				contexts[receiverid] = nil


Not passed. I am too unsure about that one too. This needs testing.

Looks correct judging from https://github.com/ShiroKSH/wire/blob/452f57412fc23530d3b2892d0ad2f4b1744e2a71/lua/entities/gmod_wire_expression2/core/signal.lua#L126

Fine, I guess.

Grocel · 2026-07-09T16:41:23Z

 		"gitrid.sh",
 		"LICENSE",
-		"wiremod.*"
+		"wiremod.*",


Grocel · 2026-07-09T16:43:06Z

+					local compressed = util.Compress(data)
 					net.WriteBool(false)
-					net.WriteUInt(#data, 12)
-					net.WriteData(data)
+					if compressed then
+						net.WriteUInt(#compressed, 12)
+						net.WriteData(compressed)
+					else
+						net.WriteUInt(0, 12)
+					end


Does the receiver reflect these changes too? I don't see it in this PR.

Grocel · 2026-07-09T16:43:23Z

+	local sent = scripted_ents.Get(Data.Class)
+	if not sent or not sent.ClassName then return false end
+
+	Data.Class = sent.ClassName


Grocel · 2026-07-09T16:46:42Z

+			local folder = fop .. "/"
+			net.WriteUInt(#folder, 16)
+			net.WriteData(folder)


How does the "other side" deal with the change? Was it somehow compensated? Can the hardcoded slash be removed or moved to the receiver?

The trailing forward-slash was getting trimmed before but now it's not.

wire/lua/entities/gmod_wire_expression2/core/files.lua

Line 447 in fd918b3

table.insert(plist.data, net.ReadData(net.ReadUInt(16)))

Should test that this change in behavior is ok.

There's another bug on line 95 I just noticed where non-txt files will cause a mismatch in amount of data written and number of files specified at line 93.

If this is tested, I would give it a pass.

Tested the file list roundtrip in GMod with file and folder entries. The folder keeps the trailing /, and the count now only includes the transmitted .txt files, so non-txt files do not shift the receiver.

@thegrb93 What do you thing? It seems to be the only point left before passing.

Grocel · 2026-07-09T16:49:40Z

-						local x,y,z = string.match( val, "^ *([^%s,]+) *, *([^%s,]+) *$" )
-						if x and y and z then return x..", "..y..", "..z end
+						local x,y = string.match( val, "^ *([^%s,]+) *, *([^%s,]+) *$" )
+						if x and y then return x..", "..y end


Grocel · 2026-07-09T16:49:59Z

-				local tempVal = castE2TypeToWireValueType["VECTOR4"](val[2], typeID(val[2]))
-				if not tempVal then return "Incorrect 'value' parameter #"..i.." value! Expected 'VECTOR2'. Got: "..tostring(val[2]) end
+				local tempVal = castE2TypeToWireValueType["VECTOR4"](val, TypeID(val))
+				if not tempVal then return "Incorrect 'value' parameter #"..i.." value! Expected 'VECTOR4'. Got: "..tostring(val[2]) end


Grocel · 2026-07-09T16:50:07Z

 				val = {wireValueType, CastFunc(val[2], TypeID(val[2]))}
 			elseif #val == 2 then -- vector2
-				local tempVal = castE2TypeToWireValueType["VECTOR2"](val[2], typeID(val[2]))
+				local tempVal = castE2TypeToWireValueType["VECTOR2"](val, TypeID(val))


Grocel · 2026-07-09T16:51:25Z

 				elseif e2TypeID == TYPE_ANGLE then val = {"ANGLE", castE2TypeToWireValueType["ANGLE"](val, e2TypeID)}
 				elseif e2TypeID == TYPE_STRING then val = {"STRING", castE2TypeToWireValueType["STRING"](val, e2TypeID)}
-				else return "Incorrect 'value' parameter #"..i.." type! Expected table (Ex. table(\"normal\", 0)). Got: "..type( steamid ) end
+				else return "Incorrect 'value' parameter #"..i.." type! Expected table (Ex. table(\"normal\", 0)). Got: "..type( val ) end


There was steamid before? Strange. Looks ok, though.

ShiroKSH · 2026-07-09T17:31:29Z

@Grocel splited, moved the passed review changes to #3650.

ShiroKSH · 2026-07-09T19:02:27Z

Pushed a small WireNet update so the handler-name update stays inline if compression fails or does not fit the 12-bit path.

Tested in GMod on gm_construct: signalSendToPlayer delivery, signal listener cleanup on chip removal, E2 file list roundtrip with file + folder entries, and WireLib.Net.Trivial normal/nil-compress/too-large-compress cases.

Should be ready for another look.

thegrb93 · 2026-07-09T19:35:39Z

This is all wrong. Please change it back to how it was in the first pr.

This gets difficult to keep track of, lol.

Grocel · 2026-07-09T20:14:39Z

 			for name, contexts in pairs_ac(signals) do
 				-- to remove all signals the chip registered for.
-				contexts[self] = nil
+				contexts[receiverid] = nil


Fine, I guess.

Grocel · 2026-07-09T20:15:17Z

-	if not IsValid(player) then return end
-	broadcastSignal(self.data.signalgroup, name, 1, self.entity, player)
+	if not IsValid(player) or not player:IsPlayer() then return end
+	broadcastSignal(self.data.signalgroup, name, 1, self.entity, player:UniqueID())


Grocel · 2026-07-09T20:16:25Z

This gets difficult to keep track of, lol.

Grocel · 2026-07-09T20:18:18Z

+			local folder = fop .. "/"
+			net.WriteUInt(#folder, 16)
+			net.WriteData(folder)


If this is tested, I would give it a pass.

Grocel · 2026-07-09T20:20:37Z


--local function table_IsEmpty(t) return not pairs(t)(t) end
-local function table_IsEmpty(t) return not next(t) end
-


Passed. But I would avoid adding more stuff during the review process. This change can stay in, though.

blueshank-gh · 2026-07-09T22:22:30Z

@ShiroKSH once you feel everything about these guys have been addressed according to Grocel, simply change this to open state, and they can have the final say on merge, which from how this looks is 100% going to happen.

ShiroKSH · 2026-07-09T22:23:56Z

once you feel everything about these guys have been addressed according to Grocel, simply change this to open state, and they can have the final say on merge, which from how this looks is 100% going to happen.

ye i knew ty <3

fix: keep reviewed wire runtime fixes

3ff85cc

ShiroKSH mentioned this pull request Jul 9, 2026

Fix Wire runtime and tool handling edge cases #3648

Closed

blueshank-gh reviewed Jul 9, 2026

View reviewed changes

fix: drop fpga operator change

a2a9973

ShiroKSH marked this pull request as draft July 9, 2026 16:18

Grocel suggested changes Jul 9, 2026

View reviewed changes

chore: split passed review fixes

dba9842

fix: keep wirenet fallback inline

5fe8d32

ShiroKSH marked this pull request as ready for review July 9, 2026 19:02

fix: remove unused signal helper

452f574

ShiroKSH requested a review from Grocel July 9, 2026 19:09

ShiroKSH marked this pull request as draft July 9, 2026 19:09

thegrb93 reviewed Jul 9, 2026

View reviewed changes

fix: align reviewed network payload changes

44fe4aa

thegrb93 reviewed Jul 9, 2026

View reviewed changes

Comment thread lua/wire/wirenet.lua Outdated

fix: drop redundant compression size check

8c1dbef

Grocel reviewed Jul 9, 2026

View reviewed changes


		--local function table_IsEmpty(t) return not pairs(t)(t) end
		local function table_IsEmpty(t) return not next(t) end

Uh oh!

Conversation

ShiroKSH commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ShiroKSH Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

blueshank-gh commented Jul 9, 2026

Uh oh!

Grocel left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Grocel Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ShiroKSH commented Jul 9, 2026

Uh oh!

ShiroKSH commented Jul 9, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

blueshank-gh commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ShiroKSH commented Jul 9, 2026 •

edited

Loading

ShiroKSH Jul 9, 2026 •

edited

Loading

Grocel Jul 9, 2026 •

edited

Loading

blueshank-gh commented Jul 9, 2026 •

edited

Loading