[Solved w/ workaround] Question about dwm-pango and statuscolors patch

lswest · 2010-05-30 13:24:25

I've recently started using the Pango patch for dwm to enable unicode characters in the statusbar. However, I ended up being unable to patch dwm-pango with the pertag and statuscolors patches, since the pango patch results in conflicts/changes that require the .diff file to be re-generated. I've created a working pertag patch for dwm-pango 5.8.1. However, the statuscolors patch isn't as easy to re-write.

I tried re-writing it, and managed to avoid any actual errors (I had trouble this this section initially:

+        dc.colors[i][ColBorder] = getcolor(colors[i][ColBorder], dc.xftnorm+colors[i][ColBorder]);
+        dc.colors[i][ColBG] = getcolor(colors[i][ColFG], dc.xftnorm+colors[i][ColFG]);
+        dc.colors[i][ColFG] = getcolor(colors[i][ColFG], dc.xftnorm+colors[i][ColFG]);

but managed to solve it by playing around with it for a bit).

However, once I compiled it the statusbar was huge (about 25 pixels instead of 12), and the colors used were black and white, which are actually the "alert" colors by default. Also, the text was padded so much that you could barely see the top of the text (sorry, didn't take a screenshot at the time).

Anyways, my question is this:
Has anyone generated a .diff file for statuscolors that works with dwm-pango? I'll give it another go in a little while, but I thought I'd ask here during my "break". If anyone is interested, my first result was this: (sorry, first .diff file I ever created, and so I ended up mixing all 3 patches together!).

Thanks in advance for any suggestions/replies,
Lswest

*EDIT* Fixed the diff, and it works now, but I have one issue that I have yet to solve. First things first, fixed diff is here: http://lswest.pastebin.com/dHnmKdkv

The problem I have now is the following:

As you can see, the foreground color isn't applied to the text when selected anymore. This is also true for the Status Bar section (where currently focused windows are shown). Anyone have a suggestion where the issue could lie?

Config.h: http://lswest.pastebin.com/AJm0VYwq

Last edited by lswest (2010-05-31 08:20:29)

Runiq · 2010-05-30 15:24:40

The problem most probably lies within the second-to-last hunk in your patch:

@@ -1499,12 +1529,11 @@ setup(void) {
     cursor[CurMove] = XCreateFontCursor(dpy, XC_fleur);
 
     /* init appearance */
-        dc.norm[ColBorder] = getcolor(normbordercolor, dc.xftnorm+ColBorder);
-        dc.norm[ColBG] = getcolor(normbgcolor, dc.xftnorm+ColBG);
-        dc.norm[ColFG] = getcolor(normfgcolor, dc.xftnorm+ColFG);
-        dc.sel[ColBorder] = getcolor(selbordercolor, dc.xftsel+ColBorder);
-        dc.sel[ColBG] = getcolor(selbgcolor, dc.xftsel+ColBG);
-        dc.sel[ColFG] = getcolor(selfgcolor, dc.xftsel+ColFG);
+    for(int i=0; i<NUMCOLORS; i++) {
+        dc.colors[i][ColBorder] = getcolor(colors[i][ColBorder], dc.xftnorm+ColBorder);
+        dc.colors[i][ColBG] = getcolor(colors[i][ColBG], dc.xftnorm+ColBG);
+        dc.colors[i][ColFG] = getcolor(colors[i][ColFG], dc.xftnorm+ColFG);
+        }
 
     dc.drawable = XCreatePixmap(dpy, root, DisplayWidth(dpy, screen), bh, DefaultDepth(dpy, screen));
     dc.gc = XCreateGC(dpy, root, 0, NULL);

You probably have to somehow adjust the "dc.xftnorm" and "dc.xftsel" depending on i. I'm trying to come up with something, but it'll probably take a while, sorry. As mentioned on another thread I'm not much of a coder.

lswest · 2010-05-30 15:29:14

Thanks for the reply

I'm (honestly) not much of a coder either I'll have a look and a bit of a play around, thanks for the suggestion and help!

Oh, and I just noticed we're both in Germany Talk about coincidence.

*EDIT*

I've played with it and changed the code to this:

    for(int i=0; i<NUMCOLORS; i++) {
            if(i == 1){
                dc.colors[i][ColBorder] = getcolor(colors[i][ColBorder], dc.xftnorm+ColBorder);
                dc.colors[i][ColBorder] = getcolor(colors[i][ColBG], dc.xftnorm+ColBG);
                dc.colors[i][ColBorder] = getcolor(colors[i][ColFG], dc.xftsel+ColFG);
            }else {
            dc.colors[i][ColBorder] = getcolor(colors[i][ColBorder], dc.xftnorm+ColBorder);
                dc.colors[i][ColBG] = getcolor(colors[i][ColBG], dc.xftnorm+ColBG);
                dc.colors[i][ColFG] = getcolor(colors[i][ColFG], dc.xftnorm+ColFG);
            }
        }

And now the result is this:

I'm on tag #2 and the little box that denotes open windows is gone, and I'm (honestly) quite lost as to whether this is a step in the right direction or the wrong one.

Last edited by lswest (2010-05-30 15:47:28)

Runiq · 2010-05-30 15:53:27

lswest wrote:

And now the result is this:
http://imgur.com/YLiyF.png
I'm on tag #2 and the little box that denotes open windows is gone, and I'm (honestly) quite lost as to whether this is a step in the right direction or the wrong one.

What's going on with tag #4?

lswest · 2010-05-30 15:56:14

Tag 4 is sending an "alert" (what statuscolors is actually good for, changes the color when there's an alert - i.e. unread message in IM, etc.), and there the text still isn't displayed (not selected though, so it may be an issue with something else).

Got it solved.

    for(int i=0; i<NUMCOLORS; i++) {
                if(i == 1) {
                   dc.colors[i][ColBorder] = getcolor(colors[i][ColBorder], dc.xftsel+ColBorder);
                   dc.colors[i][ColBG] = getcolor(colors[i][ColBG], dc.xftsel+ColBG);
                   dc.colors[i][ColFG] = getcolor(colors[i][ColFG], dc.xftsel+ColFG);
                } else {
                dc.colors[i][ColBorder] = getcolor(colors[i][ColBorder], dc.xftnorm+ColBorder);
                    dc.colors[i][ColBG] = getcolor(colors[i][ColBG], dc.xftnorm+ColBG);
                    dc.colors[i][ColFG] = getcolor(colors[i][ColFG], dc.xftnorm+ColFG);
                }
        }

Was what I needed. Now I need to figure out how to get the alert text to show up right. In any case, I'll update the thread once I figure out the alert text with a link to pastebin for the statuscolors patch for anyone interested.

*EDIT*
I haven't figured out an actual solution yet, but I have figured out a work-around (adjusted the alert colors to work with the foreground color I use by default). Diff file is here: http://lswest.pastebin.com/nLei3rfj

Not sure if I will figure out an actual solution to this problem, since I don't need anything more than a slightly different background color to see alerts.

Thanks to runiq for the suggestion that pointed me in the right direction.

Last edited by lswest (2010-05-30 16:20:30)

Runiq · 2010-05-30 16:18:39

I'm using another approach, I made an array xftcolors with the same parameters as colors. Hopefully that'll work. If it does, it should solve the box thingy too.

lswest · 2010-05-30 16:20:59

Runiq wrote:

I'm using another approach, I made an array xftcolors with the same parameters as colors. Hopefully that'll work. If it does, it should solve the box thingy too.

Haha, I just updated my post Let me know if it works though!

Runiq · 2010-05-30 17:00:44

Hah, dude, I got it! Works perfectly and as expected, though I'm not sure about the font placement (there's one bit I'll look over again). Will post after dinner.

lswest · 2010-05-30 17:02:50

Sounds great! Talk about a productive day, eh?

Enjoy your dinner (I just got back from mine actually).

Runiq · 2010-05-30 17:40:10

lswest wrote:

Sounds great! Talk about a productive day, eh?
Enjoy your dinner (I just got back from mine actually).

Heh, it would have been productive had I managed to drag my lazy ass up and do some learning.

Pastebinned here. Apply it in the order pango -> pertag -> statuscolors. I couldn't figure out how to test this thing, as piping into dwm in .xinitrc doesn't seem to work, and setting the root window's name won't escape my backslashes… If you encounter any problems, let me know.

lswest · 2010-05-30 18:01:18

Thanks for the patch. Seem to be having a few issues applying it though:

patching file dwm.c
Hunk #2 FAILED at 104.
Hunk #3 succeeded at 168 (offset -20 lines).
Hunk #4 FAILED at 705.
Hunk #5 FAILED at 722.
Hunk #6 FAILED at 749.
Hunk #7 FAILED at 770.
Hunk #8 FAILED at 791.
Hunk #9 FAILED at 835.
Hunk #10 FAILED at 1126.
Hunk #11 FAILED at 1494.
Hunk #12 FAILED at 1662.
10 out of 12 hunks FAILED -- saving rejects to file dwm.c.rej

dwm.c.rej:

--- dwm.c 2010-05-30 18:11:46.388654711 +0200
+++ dwm.c 2010-05-30 18:42:56.131801639 +0200
@@ -104,13 +105,11 @@
 
 typedef struct {
  int x, y, w, h;
- unsigned long norm[ColLast];
- unsigned long sel[ColLast];
+        unsigned long colors[MAXCOLORS][ColLast];
  Drawable drawable;
  GC gc;
 
- XftColor  xftnorm[ColLast];
- XftColor  xftsel[ColLast];
+        XftColor  xftcolors[MAXCOLORS][ColLast];
  XftDraw  *xftdrawable;
 
  PangoContext *pgc;
@@ -705,14 +705,13 @@
  dc.x = 0;
  for(i = 0; i < LENGTH(tags); i++) {
    dc.w = TEXTW(tags[i]);
-   col = m->tagset[m->seltags] & 1 << i ? dc.sel : dc.norm;
-   drawtext(tags[i], col, urg & 1 << i);
-   drawsquare(m == selmon && selmon->sel && selmon->sel->tags & 1 << i,
-              occ & 1 << i, urg & 1 << i, col);
+   col = dc.colors[ (m->tagset[m->seltags] & 1 << i ? 1:(urg & 1 << i ? 2:0))];
+   drawtext(tags[i], col, True);
+   drawsquare(m == selmon && selmon->sel && selmon->sel->tags & 1 << i, occ & 1 << i, col);
    dc.x += dc.w;
  }
  dc.w = blw = TEXTW(m->ltsymbol);
- drawtext(m->ltsymbol, dc.norm, False);
+ drawtext(m->ltsymbol, dc.colors[0], True);
  dc.x += dc.w;
  x = dc.x;
  if(m == selmon) { /* status is only drawn on selected monitor */
@@ -722,19 +721,19 @@
      dc.x = x;
      dc.w = m->ww - x;
    }
-   drawtext(stext, dc.norm, False);
+   drawcoloredtext(stext);
  }
  else
    dc.x = m->ww;
  if((dc.w = dc.x - x) > bh) {
    dc.x = x;
    if(m->sel) {
-     col = m == selmon ? dc.sel : dc.norm;
-     drawtext(m->sel->name, col, False);
-     drawsquare(m->sel->isfixed, m->sel->isfloating, False, col);
+     col = m == selmon ? dc.colors[1] : dc.colors[0];
+     drawtext(m->sel->name, col, True);
+     drawsquare(m->sel->isfixed, m->sel->isfloating, col);
    }
    else
-     drawtext(NULL, dc.norm, False);
+     drawtext(NULL, dc.colors[0], False);
  }
  XCopyArea(dpy, dc.drawable, m->barwin, dc.gc, 0, 0, m->ww, bh, 0, 0);
  XSync(dpy, False);
@@ -749,12 +748,42 @@
 }
 
 void
-drawsquare(Bool filled, Bool empty, Bool invert, unsigned long col[ColLast]) {
+drawcoloredtext(char *text) {
+ Bool first=True;
+ char *buf = text, *ptr = buf, c = 1;
+ unsigned long *col = dc.colors[0];
+ int i, ox = dc.x;
+
+ while( *ptr ) {
+   for( i = 0; *ptr < 0 || *ptr > NUMCOLORS; i++, ptr++);
+   if( !*ptr ) break;
+   c=*ptr;
+   *ptr=0;
+   if( i ) {
+     dc.w = selmon->ww - dc.x;
+     drawtext(buf, col, first);
+     dc.x += textnw(buf, i) + textnw(&c,1);
+     if( first ) dc.x += ( dc.font.ascent + dc.font.descent ) / 2;
+     first = False;
+   } else if( first ) {
+     ox = dc.x += textnw(&c,1);
+   }
+   *ptr = c;
+   col = dc.colors[ c-1 ];
+   buf = ++ptr;
+ }
+ if( !first ) dc.x-=(dc.font.ascent+dc.font.descent)/2;
+ drawtext(buf, col, True);
+ dc.x = ox;
+}
+
+void
+drawsquare(Bool filled, Bool empty, unsigned long col[ColLast]) {
  int x;
  XGCValues gcv;
  XRectangle r = { dc.x, dc.y, dc.w, dc.h };
 
- gcv.foreground = col[invert ? ColBG : ColFG];
+ gcv.foreground = col[ ColFG ];
  XChangeGC(dpy, dc.gc, GCForeground, &gcv);
  x = (dc.font.ascent + dc.font.descent + 2) / 4;
  r.x = dc.x + 1;
@@ -770,17 +799,17 @@
 }
 
 void
-drawtext(const char *text, unsigned long col[ColLast], Bool invert) {
+drawtext(const char *text, unsigned long col[ColLast], Bool pad) {
  char buf[256];
  int i, x, y, h, len, olen;
  XRectangle r = { dc.x, dc.y, dc.w, dc.h };
 
- XSetForeground(dpy, dc.gc, col[invert ? ColFG : ColBG]);
+ XSetForeground(dpy, dc.gc, col[ ColBG ]);
  XFillRectangles(dpy, dc.drawable, dc.gc, &r, 1);
  if(!text)
    return;
  olen = strlen(text);
- h = dc.font.ascent + dc.font.descent;
+ h = pad ? (dc.font.ascent + dc.font.descent) : 0;
  y = dc.y;
  x = dc.x + (h / 2);
  /* shorten text if necessary */
@@ -791,7 +820,7 @@
  if(len < olen)
    for(i = len; i && i > len - 3; buf[--i] = '.');
  pango_layout_set_text(dc.plo, text, len);
- pango_xft_render_layout(dc.xftdrawable, (col==dc.norm?dc.xftnorm:dc.xftsel)+(invert?ColBG:ColFG), dc.plo, x * PANGO_SCALE, y * PANGO_SCALE);
+ pango_xft_render_layout(dc.xftdrawable, (col==dc.colors[0]?dc.xftcolors[0]:dc.xftcolors[1])+(ColFG), dc.plo, x * PANGO_SCALE, y * PANGO_SCALE);
 }
 
 void
@@ -835,7 +864,7 @@
    detachstack(c);
    attachstack(c);
    grabbuttons(c, True);
-   XSetWindowBorder(dpy, c->win, dc.sel[ColBorder]);
+   XSetWindowBorder(dpy, c->win, dc.colors[1][ColBorder]);
    XSetInputFocus(dpy, c->win, RevertToPointerRoot, CurrentTime);
  }
  else
@@ -1126,7 +1155,7 @@
  }
  wc.border_width = c->bw;
  XConfigureWindow(dpy, w, CWBorderWidth, &wc);
- XSetWindowBorder(dpy, w, dc.norm[ColBorder]);
+ XSetWindowBorder(dpy, w, dc.colors[0][ColBorder]);
  configure(c); /* propagates border_width, if size doesn't change */
  updatesizehints(c);
  XSelectInput(dpy, w, EnterWindowMask|FocusChangeMask|PropertyChangeMask|StructureNotifyMask);
@@ -1494,12 +1523,11 @@
  cursor[CurMove] = XCreateFontCursor(dpy, XC_fleur);
 
  /* init appearance */
-        dc.norm[ColBorder] = getcolor(normbordercolor, dc.xftnorm+ColBorder);
-        dc.norm[ColBG] = getcolor(normbgcolor, dc.xftnorm+ColBG);
-        dc.norm[ColFG] = getcolor(normfgcolor, dc.xftnorm+ColFG);
-        dc.sel[ColBorder] = getcolor(selbordercolor, dc.xftsel+ColBorder);
-        dc.sel[ColBG] = getcolor(selbgcolor, dc.xftsel+ColBG);
-        dc.sel[ColFG] = getcolor(selfgcolor, dc.xftsel+ColFG);
+ for(int i=0; i<NUMCOLORS; i++) {
+   dc.colors[i][ColBorder] = getcolor( colors[i][ColBorder], dc.xftcolors[i]+ColBorder );
+   dc.colors[i][ColFG] = getcolor( colors[i][ColFG], dc.xftcolors[i]+ColFG );
+   dc.colors[i][ColBG] = getcolor( colors[i][ColBG], dc.xftcolors[i]+ColBG );
+ }
 
  dc.drawable = XCreatePixmap(dpy, root, DisplayWidth(dpy, screen), bh, DefaultDepth(dpy, screen));
  dc.gc = XCreateGC(dpy, root, 0, NULL);
@@ -1662,7 +1690,7 @@
  if(!c)
    return;
  grabbuttons(c, False);
- XSetWindowBorder(dpy, c->win, dc.norm[ColBorder]);
+ XSetWindowBorder(dpy, c->win, dc.colors[0][ColBorder]);
  XSetInputFocus(dpy, root, RevertToPointerRoot, CurrentTime);
 }

Probably solvable by just editing the .diff a bit, I'll give it a proper go tomorrow morning, fresh start and all.

Thanks again for all your help!
Lswest

And yeah...I class any day where I do more than just read/chat as productive

Runiq · 2010-05-30 21:12:57

Hunk #2 FAILED at 104.
Hunk #3 succeeded at 168 (offset -20 lines).
Hunk #4 FAILED at 705.
Hunk #5 FAILED at 722.
Hunk #6 FAILED at 749.
Hunk #7 FAILED at 770.
Hunk #8 FAILED at 791.
Hunk #9 FAILED at 835.
Hunk #10 FAILED at 1126.
Hunk #11 FAILED at 1494.
Hunk #12 FAILED at 1662.
10 out of 12 hunks FAILED -- saving rejects to file dwm.c.rej

Argh, not again. That's just my rotten luck. I've had the same problem with a patch I created in the libetc thread. I'll look over it once again and mail it to you if nothing else helps.

Last edited by Runiq (2010-05-30 21:13:29)

lswest · 2010-05-31 08:16:44

I've fixed the .diff file, and it patches properly, but the blue colour still isn't drawn into the white square on alerts. Works fine if you use a background colour on which you can read the text's usual colour. I think it may just be the way DWM draws the box when you apply statuscolors on it. I'll bet you'd need to set dc.xftdrawable when changing the colour or so, but it's really not the end of the world for me, so I think I'll just leave it where it is now.

Diff file: http://lswest.pastebin.com/97E87UUC

Thanks again for all your help!
Lswest

And the issue with the patch was simply that some of the lines that told it where to make edits weren't an exact match/weren't where it thought they were. Also, there were a few lines you missed out that needed changes. If you want to know how I fixed it, I simply extracted the dwm-5.8.1 source, patched it with pango, then pertag, and then I manually made the changes from your .diff, and then fixed the errors it threw out during compile. To create the diff file I just extracted the source again, patched it with pango and pertag, and ran

diff -up dwm-5.8.1 dwm-5.8.1-patched > dwm-pango-5.8.1-statuscolors.diff

Maybe that'll help you sort out the libetc issue?

*EDIT* I've been thinking about it a bit more, and I have a feeling the problem lies with the way dwm assigns the foreground colours when patched with pango (at least...I think dwm-5.7.2 without pango drew the text properly), and that it simply never assigns the foreground colour from the array for alerts (it would be colors[2]). Not sure how to check or fix it, but I figured I should stick the thought here in case it spurs someone in the right direction.

Last edited by lswest (2010-05-31 08:43:05)

Aakko · 2010-06-20 11:25:05

Is there a working statuscolors patch for dwm 5.8.2 yet?

lswest · 2010-06-20 13:26:36

If the patch fails, you can make the changes by hand and generate a diff like that. If there's no 5.8.2 patch on the dwm website, there's either a) no need to change the diff file, or b) simply no patch yet.

I generated a diff file for dwm-pango 5.8.2, so if you want that, it's here: http://lswest.pastebin.com/tJAMsMcq

Hope that helps,
Lswest

cirnOS · 2011-11-04 02:00:05

I'd post mine but I'm currently working on applying patches to DWM and changing the colors. I haven't gotten mine setup yet however.

Last edited by cirnOS (2011-11-04 03:04:46)

Arch Linux

#1 2010-05-30 13:24:25

[Solved w/ workaround] Question about dwm-pango and statuscolors patch

#2 2010-05-30 15:24:40

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#3 2010-05-30 15:29:14

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#4 2010-05-30 15:53:27

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#5 2010-05-30 15:56:14

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#6 2010-05-30 16:18:39

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#7 2010-05-30 16:20:59

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#8 2010-05-30 17:00:44

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#9 2010-05-30 17:02:50

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#10 2010-05-30 17:40:10

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#11 2010-05-30 18:01:18

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#12 2010-05-30 21:12:57

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#13 2010-05-31 08:16:44

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#14 2010-06-20 11:25:05

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#15 2010-06-20 13:26:36

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

#16 2011-11-04 02:00:05

Re: [Solved w/ workaround] Question about dwm-pango and statuscolors patch

Board footer