On Sun, 11 Aug 2024 07:26:16 -0400, Chris Ahlstrom wrote:

> In the deep past I tried ctags, but it never took hold with me. I see
> there's a thing called LSP (Language Server Protocol).

I've used ctags and csope and have a slight preference for cscope. When
the codebase contains a few hundred .c files it help to be able to work
your way through the labyrinth. cscope allows you to find instances of
where a function is used. iirc ctags only works back through definitions.

On 11 Aug 2024 17:30:31 GMT, rbowman wrote:

> I've used ctags and csope and have a slight preference for cscope.

I tried etags for Emacs, then wondered why no one had the bright idea of
storing the symbol definitions in an SQLite database for quick lookup.

So I created project-tags <https://gitlab.com/ldo/emacs-prefs>.

rbowman wrote this copyrighted missive and expects royalties:

> On Sun, 11 Aug 2024 07:26:16 -0400, Chris Ahlstrom wrote:
>
>> In the deep past I tried ctags, but it never took hold with me. I see
>> there's a thing called LSP (Language Server Protocol).
>
> I've used ctags and csope and have a slight preference for cscope. When
> the codebase contains a few hundred .c files it help to be able to work
> your way through the labyrinth. cscope allows you to find instances of
> where a function is used. iirc ctags only works back through definitions.

These days I use grep to find calls or definitions. I use the AT&T style, at
least the part where the return type is on one line and the function name is on
the next line. Function calls are for the form func() while declaration and
definition of of the form func ().

If there are many parameters, I do not indent the 2nd and above parameters to
line up with the first parameter. I do this:

func
( bool parameter1,
bool parameter2,
bool parameter
) {
... code
}

Similar with printf():

printf
( "long format statement part 1"
"long format statement part 2"
,
list of parameters to print
);

Anything to keep lines under 78 characters and be able to scan straight down
with my eyes.

OCD?

--
Q: How many lawyers does it take to change a light bulb?
A: Whereas the party of the first part, ... (fortune toooo looonnnng)

Lawrence D'Oliveiro wrote this copyrighted missive and expects royalties:

> On 11 Aug 2024 17:30:31 GMT, rbowman wrote:
>
>> I've used ctags and csope and have a slight preference for cscope.
>
> I tried etags for Emacs, then wondered why no one had the bright idea of
> storing the symbol definitions in an SQLite database for quick lookup.
>
> So I created project-tags <https://gitlab.com/ldo/emacs-prefs>.

Cool!

I'd probably use std::unordered_multimap though.

--
You look tired.

On Mon, 12 Aug 2024 07:29:36 -0400, Chris Ahlstrom wrote:

> These days I use grep to find calls or definitions. I use the AT&T
> style, at least the part where the return type is on one line and the
> function name is on the next line. Function calls are for the form
> func() while declaration and definition of of the form func ().

I do similar:

void*
send_to_cjin(void* arg)
{ unsigned short id = 1;
char outgoing[32000];
char* query = NULL;
char msg[1024];
char* query_response = NULL;
char label[256] = "";
int ret;
size_t outgoing_length;

I used to be more anal about lining stuff up nicely. That was more from
'when in Rome' when working with existing code. We didn't have a style
guide so the legacy stuff was a mixed bag. Somewhere along the line I
came to prefer 'char* query' rather than 'char *query'. A friend
disagreed but we seldom worked on the same code.

I sometimes use grep as in

find . -name "*.c" -print | xargs grep -il pthread_cond_wait

There is another form that uses exec and {} that I seldom remember. Most
often I used a Python script but I only got around to converting it to 3.x
recently.

rbowman wrote this copyrighted missive and expects royalties:

> On Mon, 12 Aug 2024 07:29:36 -0400, Chris Ahlstrom wrote:
>
>> These days I use grep to find calls or definitions. I use the AT&T
>> style, at least the part where the return type is on one line and the
>> function name is on the next line. Function calls are for the form
>> func() while declaration and definition of of the form func ().
>
> I do similar:
>
> void*
> send_to_cjin(void* arg)
> {
> unsigned short id = 1;
> char outgoing[32000];
> char* query = NULL;
> char msg[1024];
> char* query_response = NULL;
> char label[256] = "";
> int ret;
> size_t outgoing_length;
>
> I used to be more anal about lining stuff up nicely. That was more from
> 'when in Rome' when working with existing code. We didn't have a style
> guide so the legacy stuff was a mixed bag. Somewhere along the line I
> came to prefer 'char* query' rather than 'char *query'. A friend
> disagreed but we seldom worked on the same code.

I put spaces around both sides of "*", "&", and "[]" in declarations.

char * query = nullptr;
char msg [1024];
const std::string & errmsg = error_message();

But then:

char first = &msg[0];
std::string msg2 = *msgstring;

> I sometimes use grep as in
>
> find . -name "*.c" -print | xargs grep -il pthread_cond_wait
>
> There is another form that uses exec and {} that I seldom remember. Most
> often I used a Python script but I only got around to converting it to 3.x
> recently.

And for editing (in a bash terminal window):

vi $(find . -name "*.c" -print | xargs grep -il pthread_cond_wait)

Anyway, our group created a coding-conventions document, but everyone did
it the way they wanted to anyway. A herd o' cats.

Or you can lie outrageously and frequently:

--
If you tell the truth you don't have to remember anything.
-- Mark Twain

On Tue, 13 Aug 2024 07:24:29 -0400, Chris Ahlstrom wrote:

> Anyway, our group created a coding-conventions document, but everyone
> did it the way they wanted to anyway. A herd o' cats.

We let most personal quirks slide but people who insisted on putting
multiple statements on one line tended to get beaten up in meetings
(figuratively) except for uses like

if (foo == bar) baz = 1;

rbowman wrote this copyrighted missive and expects royalties:

> On Tue, 13 Aug 2024 07:24:29 -0400, Chris Ahlstrom wrote:
>
>> Anyway, our group created a coding-conventions document, but everyone
>> did it the way they wanted to anyway. A herd o' cats.
>
> We let most personal quirks slide but people who insisted on putting
> multiple statements on one line tended to get beaten up in meetings
> (figuratively) except for uses like
>
> if (foo == bar) baz = 1;

We had one guy who would nest the "?:" operator. Almost impossible to grok with
4 or more nestings.

a ? b : c ? d : e ? f : g ? h : i

--
Q: How many lawyers does it take to change a light bulb?
A: You won't find a lawyer who can change a light bulb. Now, if
you're looking for a lawyer to screw a light bulb...

On Wed, 14 Aug 2024 16:42:58 -0400, Chris Ahlstrom wrote:

> We had one guy who would nest the "?:" operator. Almost impossible to
> grok with 4 or more nestings.
>
> a ? b : c ? d : e ? f : g ? h : i

I have seen that -- and it's always the same guy who thinks more than one
letter for a variable is a waste. That guy would make an exception for
arrays. They were 'ary'.

Funny thing was when he was a TA people who were in his section said he
was a real hard ass for comments and meaningful names. When he became
gainfully employed he never used either and constructed macros that would
make K&R cry.

rbowman wrote this copyrighted missive and expects royalties:

> On Wed, 14 Aug 2024 16:42:58 -0400, Chris Ahlstrom wrote:
>
>> We had one guy who would nest the "?:" operator. Almost impossible to
>> grok with 4 or more nestings.
>>
>> a ? b : c ? d : e ? f : g ? h : i
>
> I have seen that -- and it's always the same guy who thinks more than one
> letter for a variable is a waste. That guy would make an exception for
> arrays. They were 'ary'.

Actually that was just an example. However, I will use very short variable
names if the context is obvious:

for (auto c : a_string_variable) { do something with c }
for (char c : a_string_variable) { do something with c }

And never camel-case; I hate that.

> Funny thing was when he was a TA people who were in his section said he
> was a real hard ass for comments and meaningful names. When he became
> gainfully employed he never used either and constructed macros that would
> make K&R cry.

You want a good cry? Look through the GNU gettext library code. Jeezuz!

--
Q: How many Bell Labs Vice Presidents does it take to change a light bulb?
A: That's proprietary information. Answer available from AT&T on payment
of license fee (binary only).

On Thu, 15 Aug 2024 06:37:28 -0400, Chris Ahlstrom wrote:

> Actually that was just an example. However, I will use very short
> variable names if the context is obvious:
>
> for (auto c : a_string_variable) { do something with c }
> for (char c : a_string_variable) { do something with c }
>
> And never camel-case; I hate that.

Yeah. I was warped by FORTRAN at an early age so I do use i, j, and k a
lot.

I sometimes use CamelCase but try to avoid Hungarian notation unless I'm
in legacy C++ code.

When I was playing with Pyside6 in Python there is a directive that let's
you use the snakecase equivalent. However it doesn't work for everything
and rather than try to find out which places it didn't and wind up with a
mixed mess I went to the Qt conventions.

Re: if (foo == bar) baz = 1;

baz = foo == bar ;

rBowman: We let most personal quirks slide but people who insisted on putting
multiple statements on one line tended to get "beaten up" in meetings.

From "z1.CPP" in "Jeff-Relf.Me/z1.ZIP": // "Comic Code" cost me 35 $

CharH = 0, CharW = WinW/ScrColumns ;
Font( FixedWFont, L"Comic Code" ), !FixedWFont.Draw && Font( FixedWFont, L"Consolas" ), Font( Symbols, L"Segoe UI Symbol" );
Font( SegoeUI, L"Segoe UI" ), Font( YaHei, L"Microsoft YaHei" );

iCharH = CharH, FontCacheW = WinW, IncCache, WinR = rn( WinH / iCharH ) - RowsTop ;

float cw, _CharW = CharW, _CharH = CharH ; MarioWinW = ( WinW < WinH ? WinW : WinW / 2 ), cw = MarioWinW / 80 ;
ChessLeft = MonplyLeft = MarioLeft = cw, MarioWinW -= 2*MarioLeft, ScaMonply = MarioWinW / 2500 ;
CharW = cw, Font( MonoplyFont, L"Segoe Print" ), gCharW = CharW, MonopCharH = GlyphH * 1.5 ;
CharW = cw * 2.2, Font( SolitaireFont, L"Segoe Print" ), CharW_Sol = CharW, CharH_Sol = GlyphH * 1.5 ;
CharW = cw * 4.5, Font( ChessFont, L"Segoe UI Symbol" ), CharW_Chess = 1.7 * CharW, CharH_Chess = GlyphH ;
CharW = .9*_CharW, Font( Emoji, L"Segoe UI Emoji" );

On Fri, 16 Aug 2024 06:06:46 -0700 (Seattle), Relf wrote:

> Re: if (foo == bar) baz = 1;
>
> baz = foo == bar ;
>
> rBowman: We let most personal quirks slide but people who insisted on
> putting
> multiple statements on one line tended to get "beaten up" in
> meetings.
>
> From "z1.CPP" in "Jeff-Relf.Me/z1.ZIP": // "Comic Code" cost me 35 $
>
> CharH = 0, CharW = WinW/ScrColumns ;
> Font( FixedWFont, L"Comic Code" ), !FixedWFont.Draw && Font(
> FixedWFont, L"Consolas" ), Font( Symbols, L"Segoe UI Symbol" );
> Font( SegoeUI, L"Segoe UI" ), Font( YaHei, L"Microsoft YaHei" );
>
> iCharH = CharH, FontCacheW = WinW, IncCache, WinR = rn( WinH / iCharH
> ) - RowsTop ;
>
> float cw, _CharW = CharW, _CharH = CharH ; MarioWinW = ( WinW <
> WinH ? WinW : WinW / 2 ), cw = MarioWinW / 80 ; ChessLeft = MonplyLeft
> = MarioLeft = cw, MarioWinW -= 2*MarioLeft, ScaMonply = MarioWinW /
> 2500 ;
> CharW = cw, Font( MonoplyFont, L"Segoe Print" ), gCharW = CharW,
> MonopCharH = GlyphH * 1.5 ;
> CharW = cw * 2.2, Font( SolitaireFont, L"Segoe Print" ), CharW_Sol =
> CharW, CharH_Sol = GlyphH * 1.5 ;
> CharW = cw * 4.5, Font( ChessFont, L"Segoe UI Symbol" ), CharW_Chess =

Introducing something like that into the code base might be good for more
than a figurative beating.

> 1.7 * CharW, CharH_Chess = GlyphH ;
> CharW = .9*_CharW, Font( Emoji, L"Segoe UI Emoji" );

You (rBowman) replied ( to me ):
> > From "z1.CPP" in "Jeff-Relf.Me/z1.ZIP": // "Comic Code" cost me 35 $
> > !Font( FixedWFont, L"Comic Code" ) && Font( FixedWFont, L"Consolas" ), Font( Symbols, L"Segoe UI Symbol" );
>
> Introducing something like that into the code base
> might be good for more than a figurative beating.

You'all would be mad, But I'd be smiling.

On Fri, 16 Aug 2024 18:43:58 -0700 (Seattle), Relf wrote:

> You (rBowman) replied ( to me ):
>> > From "z1.CPP" in "Jeff-Relf.Me/z1.ZIP": // "Comic Code" cost me 35
>> > $
>> > !Font( FixedWFont, L"Comic Code" ) && Font( FixedWFont, L"Consolas"
>> > ), Font( Symbols, L"Segoe UI Symbol" );
>>
>> Introducing something like that into the code base might be good for
>> more than a figurative beating.
>
> You'all would be mad, But I'd be smiling.

All the way to the unemployment office...

You (rBowman) replied ( to me ):
> > You'all would be mad (at my source code), But I'd be smiling.
>
> All the way to the unemployment office...

I've never received a dime of unemployment.

The pandemic destroyed me but, before that, I had the best jobs.

Harris Vows To NC Voters That She Will "Fight" For People
Being "Exploited In The Housing Market" -- YTt5e159vVk

On Fri, 16 Aug 2024 21:05:02 -0700 (Seattle), Relf wrote:

> You (rBowman) replied ( to me ):
>> > You'all would be mad (at my source code), But I'd be smiling.
>>
>> All the way to the unemployment office...
>
> I've never received a dime of unemployment.

I did receive unemployment in '72 iirc. It was summertime and the living
was easy. I came home to find a note on my door from my former boss saying
another company I had done some work for wanted to hire me. Oh, shit, back
to work. At least they could have waited until fall.

Other than that I have spent about a total of three years not working but
I wasn't unemployed. I also spent some if that time living in my pickup
but I wasn't homeless either. I had a mailing address in Sparks NV that
would forward my mail to wherever I was. Call it a sabbatical.

> Harris Vows To NC Voters That She Will "Fight" For People Being
> "Exploited In The Housing Market" -- YTt5e159vVk

Yeah, and she'll institute price controls on food. Nixon put proce
controls in place. When they unraveled poor old Jimmy Carter inherited a
real bag of shit and almost a decade of stagflation. The Kamel isn't too
good on economics or history.

a ? b : c ? d : e ? f : g ? h : i // can also be written:
a ? b : ( c ? d : ( e ? f : ( g ? h : i ) )

a ? b ? c ? d ? e : f : g : h : i // can also be written:
a ? ( b ? ( c ? ( d ? e : f ) : g ) : h ) : i

a ? a += 5, b ? b += 5, c ? c += 5, d ? d += 5, e : f : g : h : i // can also be written:
a ? ( a += 5, b ? ( b += 5, c += 5, c ? ( d ? d += 5, e : f ) : g ) : h ) : i

rBowman: [He] constructed macros that would make K&R cry.

My macros would make him cry.

On Fri, 16 Aug 2024 22:37:02 -0700 (Seattle), Relf wrote:

> a ? b : c ? d : e ? f : g ? h : i // can also be written:
> a ? b : ( c ? d : ( e ? f : ( g ? h : i ) )
>
> a ? b ? c ? d ? e : f : g : h : i // can also be written:
> a ? ( b ? ( c ? ( d ? e : f ) : g ) : h ) : i
>
> a ? a += 5, b ? b += 5, c ? c += 5, d ? d += 5, e : f : g : h : i //
> can also be written:
> a ? ( a += 5, b ? ( b += 5, c += 5, c ? ( d ? d += 5, e : f ) : g ) : h
> ) : i
>
> rBowman: [He] constructed macros that would make K&R cry.
>
> My macros would make him cry.

So where is the macro? What does this do?

#define CREATE_FMT_STR_DECL(obj) \
char *create_##obj##_fmt_str (fmt_strf_map_t *fmt)

rBowman: What does this do?
> #define CREATE_FMT_STR_DECL(obj) char *create_##obj##_fmt_str (fmt_strf_map_t *fmt)

CREATE_FMT_STR_DECL(int) // can also be written:
char *create_int_fmt_str( fmt_strf_map_t *fmt )

> > My macros would make him cry.
>
> So where is the macro?

From "z1.CPP" in "Jeff-Relf.Me/z1.ZIP":

#define g( Top, Rtn ) { while( --p##Top > Top && !*p##Top ); if ( F = p##Top + 1, p##Top <= Top ) Rtn ; \
LOOP { Cnt = SumCols = SumHrs = zCol = 0 ; \
while ( --F >= Top + 3 && ( *F && (Cnt++, SumCols += *F, SumHrs += ToHr( *F ) ) || F[-1] || F[-2] || F[-3] ) ); \
if ( F = p##Top--, Cnt >= 2 || p##Top <= Top ) break ; } if ( !Cnt ) Rtn ; \
p##Top -= rn( Trim*_DaysInCycle ), zCol = SumCols/Cnt, zHr = SumHrs/Cnt, ##Top##Pnt = { 0, 0, Color, zCol, zHr }; }

rbowman wrote this copyrighted missive and expects royalties:

> Other than that I have spent about a total of three years not working but
> I wasn't unemployed. I also spent some if that time living in my pickup
> but I wasn't homeless either. I had a mailing address in Sparks NV that
> would forward my mail to wherever I was. Call it a sabbatical.

https://www.youtube.com/watch?v=uZRN8MQA7YQ

De Phazz -- Sabbatical

I'm on permanent sabbatical... unless somebody hits me up with a good
C++/Linux gig.

--
Your lucky number is 3552664958674928. Watch for it everywhere.

On Fri, 16 Aug 2024 23:46:47 -0700 (Seattle), Relf wrote:

> From "z1.CPP" in "Jeff-Relf.Me/z1.ZIP":

You're right. I forget if it was K or R that said if he knew what pervered
things people would do with macros they wouldn't be in the language.

On Sat, 17 Aug 2024 09:53:28 -0400, Chris Ahlstrom wrote:

> I'm on permanent sabbatical... unless somebody hits me up with a good
> C++/Linux gig.

I put my two hours in this week... One of our clients edited their geocode
and I had to make sure it still worked before it went into production. HR
gets nervous if I don't have a few hours per pay period.

On 17 Aug 2024 18:56:22 GMT, rbowman wrote:

> I forget if it was K or R that said if he knew what
> pervered things people would do with macros they wouldn't be in the
> language.

What would they have instead?