doc_viewer.v C compilation error

[mike-ward]

Describe the bug

Fix: doc_viewer.v C compilation error

Problem

V compiler generates invalid C code when an inline if expression contains a function call
that uses an optional type (?Color). The C ternary can't contain variable declarations.

Reproduction Steps

• checkout commit 9077655 from https://github.com/vlang/gui
• v run examples/doc_viewer.v

Expected Behavior

Should run,

Current Behavior

Produces C Compiler Error

Possible Solution

The description above from Claude Code describes the issue.

Claude worked around the issue in doc_viewer.v in commit fcb25cf by refactoring the inline if expression.

Additional Information/Context

No response

V version

V 0.5.0 8ca3f27

Environment details (OS name and version, etc.)

V full version	V 0.5.0 e563c8f.8ca3f27
OS	macos, macOS, 26.2, 25C56
Processor	8 cpus, 64bit, little endian, Apple M2
Memory	0.16GB/8GB
V executable	/Users/mike/Documents/github/v/v
V last modified time	2026-01-30 23:25:00
V home dir	OK, value: /Users/mike/Documents/github/v
VMODULES	OK, value: /Users/mike/.vmodules
VTMP	OK, value: /tmp/v_501
Current working dir	OK, value: /Users/mike/Documents/github/gui/examples
env VFLAGS	"-message-limit 5"
env LDFLAGS	"-L/opt/homebrew/opt/ruby/lib"
Git version	git version 2.52.0
V git status	8ca3f27
.git/config present	true
cc version	Apple clang version 17.0.0 (clang-1700.6.3.2)
gcc version	Apple clang version 17.0.0 (clang-1700.6.3.2)
clang version	Apple clang version 17.0.0 (clang-1700.6.3.2)
tcc version	tcc version 0.9.28rc 2024-02-05 HEAD@105d70f7 (AArch64 Darwin)
tcc git status	thirdparty-macos-arm64 1867108f
emcc version	N/A
glibc version	N/A

Note

You can use the 👍 reaction to increase the issue's priority for developers.

Please note that only the 👍 reaction to the issue itself counts as a vote.
Other reactions and those to comments will not be taken into account.

[spytheman]

current state:

0[13:40:07]@m1: (master) /opt/v $ cd ~/.vmodules/gui
0[13:40:11]@m1: (main) ~/.vmodules/gui $ git checkout 9077655
remote: Enumerating objects: 1, done.
remote: Counting objects: 100% (1/1), done.
remote: Total 1 (delta 0), reused 1 (delta 0), pack-reused 0 (from 0)
Получаване на обекти: 100% (1/1), 1.52 KiB | 1.52 MiB/s, действието завърши.
Бележка: преминаване към „9077655“.

Указателят „HEAD“ не е свързан.  Може да разглеждате, да правите произволни
промѐни и да ги подавате.  Ако изтеглите нещо друго, всички промѐни ще се
забравят и никой клон няма да се промѐни.

Ако искате да създадете нов клон, за да запазите подаванията си, може да
направите това като зададете име на клон към опцията „-c“ на командата
„switch“.  Например:

    git switch -c ИМЕ_НА_НОВ_КЛОН

Може да отмените това действие с командата:

    git switch -

Може да спрете това съобщение със задаване на настройката „advice.detachedHead“
да е „false“ (лъжа̀).
Указателят „HEAD“ в момента сочи към 9077655 view_table: add column alignment support
0[13:40:19]@m1: ((указателят „HEAD“ не е свързан и е при „9077655“)) ~/.vmodules/gui $ v run examples/doc_viewer.v
================== C compilation error (from cc): ==============
cc:        |                  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc: /tmp/v_501/doc_viewer.01KGCG235D97SJQWDGFJPKV9FW.tmp.c:39680:72: error: expected ')'
cc:  39680 |         Array_gui__View content = (app->markdown_mode ? (               _option_gui__Color _t2 = {0};
cc:        |                                                                                            ^
cc: /tmp/v_501/doc_viewer.01KGCG235D97SJQWDGFJPKV9FW.tmp.c:39680:50: note: to match this '('
cc:  39680 |         Array_gui__View content = (app->markdown_mode ? (               _option_gui__Color _t2 = {0};
cc:        |                                                         ^
cc: /tmp/v_501/doc_viewer.01KGCG235D97SJQWDGFJPKV9FW.tmp.c:39680:81: error: expected expression
cc:  39680 |         Array_gui__View content = (app->markdown_mode ? (               _option_gui__Color _t2 = {0};
cc:        |                                                                                                     ^
cc: /tmp/v_501/doc_viewer.01KGCG235D97SJQWDGFJPKV9FW.tmp.c:39680:81: error: expected ':'
cc:  39680 |         Array_gui__View content = (app->markdown_mode ? (               _option_gui__Color _t2 = {0};
...
cc: 115 warnings and 7 errors generated.
(note: the original output was 462 lines long; it was truncated to its first 12 lines + the last line)
================================================================
Try passing `-g` when compiling, to see a .v file:line information, that correlates more with the C error.
(Alternatively, pass `-show-c-output`, to print the full C error message).
1[13:40:27]@m1: ((указателят „HEAD“ не е свързан и е при „9077655“)) ~/.vmodules/gui $

[spytheman]

Claude worked around the issue in doc_viewer.v in commit fcb25cf by refactoring the inline if expression.

Imho just make a PR with it.

[spytheman]

fcb25cf is from 2023 , imho use the full hash or link to the claude change, if you did not mean it ?

[spytheman]

vlang/gui@fcb25cf is the claude commit in vlang/gui .

[spytheman]

A more minimal case is:

pub struct Color {
pub mut:
	r u8
	g u8
	b u8
	a u8 = 255
}

pub const color_transparent = Color{0, 0, 0, 0}

pub struct MarkdownStyle {
pub:
	table_row_alt ?Color
}

pub struct MarkdownCfg {
pub:
	id    string
	style MarkdownStyle
}

struct AResult {
}

struct App {
	mode bool
}

fn (app App) method(params MarkdownCfg) AResult {
	return AResult{}
}

fn another_fn() AResult {
	return AResult{}
}

fn main() {
	app := App{}
	x := if app.mode {
		[app.method(id: 'abc')]
	} else {
		[another_fn()]
	}

	dump(x)
}

the table_row_alt ?Color is important; the rest is scaffolding to reproduce the exact failure

[mike-ward]

Apologies for the awkward bug report. I was hot and heavy into changes, and a cgen error was the least of my worries. I was going to ignore it and move on, but then I backtracked later after it was fixed to report it. Product of my impatience.

[spytheman]

Apologies for the awkward bug report. I was hot and heavy into changes, and a cgen error was the least of my worries. I was going to ignore it and move on, but then I backtracked later after it was fixed to report it. Product of my impatience.

No worries. Thank you for reporting it 🙇🏻‍♂️ .