Turns out this is not a bug! The Tokenizer counts the tabs like the
following:
else if (current_char_ == '\t') {
column_ += kTabWidth - column_ % kTabWidth;
}
where kTabWidth is 8. I was aware that Protobuf documentation recommend
indenting by 2 but I didn't know why. Now, I know.
Thank you anyway.
On Friday, May 26, 2023 at 9:47:49 AM UTC+8 Clément Jean wrote:
> I seems that the test SourceInfoTest, ScopedOptions in
> google/protobuf/compiler/parser_unittest.cc checks the case of options.
>
> I also checked the Spans there and they seem to be correct. For:
>
> message Foo {
> $a$option mopt = 1;$b$
> }
> enum Bar {
> $c$option eopt = 1;$d$
> }
> service Baz {
> $e$option sopt = 1;$f$
> rpc M(X) returns(Y) {
> $g$option mopt = 1;$h$
> }
> rpc MS4($1$stream$2$ X) returns($3$stream$4$ Y) {
> $k$option mopt = 1;$l$
> }
> }
>
> We get:
>
> SPAN: a:b
> -> 1
> -> 2
> -> 18
> SPAN: c:d
> -> 4
> -> 2
> -> 18
> SPAN: e:f
> -> 7
> -> 2
> -> 18
> SPAN: g:h
> -> 9
> -> 4
> -> 20
> SPAN: k:l
> -> 12
> -> 4
> -> 20
> SPAN: 1:2
> -> 11
> -> 10
> -> 16
> SPAN: 3:4
> -> 11
> -> 28
> -> 34
> On Friday, May 26, 2023 at 8:51:03 AM UTC+8 Clément Jean wrote:
>
>> Should I create an issue on Github for this? Is there any way I can
>> contribute to make things go faster?
>>
>> On Friday, May 26, 2023 at 2:07:17 AM UTC+8 Adam Cozzette wrote:
>>
>>> I'm not sure what's going wrong, but I agree with you that those spans
>>> in red don't look right. This might be a bug in the code that generates the
>>> source code info.
>>>
>>> On Mon, May 22, 2023 at 7:54 PM Clément Jean <[email protected]>
>>> wrote:
>>>
>>>> Also to be complete, I use the following command to turn the proto file
>>>> into a desc file:
>>>>
>>>> protoc --include_source_info --descriptor_set_out=test.desc enum.proto
>>>>
>>>> and then to have a human readable format, I run:
>>>>
>>>> cat test.desc | protoc --decode=google.protobuf.FileDescriptorSet
>>>> -I/usr/local/include/google/protobuf
>>>> /usr/local/include/google/protobuf/descriptor.proto
>>>>
>>>> finally, protoc --version returns:
>>>>
>>>> libprotoc 23.1
>>>> On Tuesday, May 23, 2023 at 8:59:21 AM UTC+8 Clément Jean wrote:
>>>>
>>>>> I'm working a parser for Protobuf and recently I am working with
>>>>> SourceLocationInfo.
>>>>> I've been noticing something weird when using options. However, I'm
>>>>> not sure if my understanding is not complete or if its a bug.
>>>>>
>>>>> I have the following enum:
>>>>>
>>>>> enum Test {
>>>>> option deprecated = true;
>>>>>
>>>>> TEST_UNSPECIFIED = 0;
>>>>> }
>>>>>
>>>>> And I get the following FileDescriptorSet (simplified):
>>>>>
>>>>> file {
>>>>> ...
>>>>> source_code_info {
>>>>> location {
>>>>> span: 0
>>>>> span: 0
>>>>> span: 4
>>>>> span: 1
>>>>> }
>>>>> location {
>>>>> path: 5
>>>>> path: 0
>>>>> span: 0
>>>>> span: 0
>>>>> span: 4
>>>>> span: 1
>>>>> }
>>>>> location {
>>>>> path: 5
>>>>> path: 0
>>>>> path: 1
>>>>> span: 0
>>>>> span: 5
>>>>> span: 9
>>>>> }
>>>>> location {
>>>>> path: 5
>>>>> path: 0
>>>>> path: 3
>>>>> span: 1
>>>>> span: 8
>>>>> span: 33
>>>>> }
>>>>> location {
>>>>> path: 5
>>>>> path: 0
>>>>> path: 3
>>>>> path: 3
>>>>> span: 1
>>>>> span: 8
>>>>> span: 33
>>>>> }
>>>>> location {
>>>>> path: 5
>>>>> path: 0
>>>>> path: 2
>>>>> path: 0
>>>>> span: 3
>>>>> span: 8
>>>>> span: 29
>>>>> }
>>>>> location {
>>>>> path: 5
>>>>> path: 0
>>>>> path: 2
>>>>> path: 0
>>>>> path: 1
>>>>> span: 3
>>>>> span: 8
>>>>> span: 24
>>>>> }
>>>>> location {
>>>>> path: 5
>>>>> path: 0
>>>>> path: 2
>>>>> path: 0
>>>>> path: 2
>>>>> span: 3
>>>>> span: 27
>>>>> span: 28
>>>>> }
>>>>> }
>>>>> }
>>>>>
>>>>> I'm confused by the ones I wrote in red. These doesn't seem to be
>>>>> correct.
>>>>> An example is span: 1 span: 8 span: 33. To the best of my knowledge,
>>>>> this means
>>>>> that we have an element at line 1 (2 in an IDE) column 8 (9 in IDE)
>>>>> that finishes on the same line (third span omitted) and at column 33.
>>>>> However, the option line is only 27 characters long...
>>>>>
>>>>> Is my mental model about SourceCodeInfo wrong? Or is it a bug?
>>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups "Protocol Buffers" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to [email protected].
>>>> To view this discussion on the web visit
>>>> https://groups.google.com/d/msgid/protobuf/a2ce2413-325d-4609-9150-804baa16d177n%40googlegroups.com
>>>>
>>>> <https://groups.google.com/d/msgid/protobuf/a2ce2413-325d-4609-9150-804baa16d177n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>> .
>>>>
>>>
--
You received this message because you are subscribed to the Google Groups
"Protocol Buffers" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/protobuf/a0eeb2e4-c294-4d82-8414-20908abe9eaen%40googlegroups.com.
