git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH (GITK) v3 2/4] gitk: Enhance file encoding support.

From
Alexander Gavrilov <angavrilov@gmail.com>
Date
Oct 13, 2008, 08:12 UTC
Message-ID
<1223885554-27718-3-git-send-email-angavrilov@gmail.com>
In-Reply-To
<1223885554-27718-2-git-send-email-angavrilov@gmail.com>

The previous patch has added functions that implement per-file contents encoding selection logic. This one modifies core gitk code to use them.

Since diffs usually contain multiple files, which may have different values of the encoding attribute, they are initially read as binary, and encoding conversion is applied to individual lines, based on the active set of diff headers.

As a side job, this patch also fixes some bugs in handling of non-ASCII filenames. Core git apparently supports only locale-encoded filenames, so processing is done using the system encoding.

Signed-off-by: Alexander Gavrilov <angavrilov@gmail.com>
Tested-by: Johannes Sixt <johannes.sixt@telecom.at>
---
 gitk |   35 +++++++++++++++++++++++++++--------
 1 files changed, 27 insertions(+), 8 deletions(-)
diff --git a/gitk b/gitk
index 8682838..5f35f61 100755
--- a/gitk
+++ b/gitk
@@ -6229,7 +6229,7 @@ proc gettree {id} {
 	    set treepending $id
 	    set treefilelist($id) {}
 	    set treeidlist($id) {}
-	    fconfigure $gtf -blocking 0
+	    fconfigure $gtf -blocking 0 -encoding binary
 	    filerun $gtf [list gettreeline $gtf $id]
 	}
     } else {
@@ -6251,11 +6251,12 @@ proc gettreeline {gtf id} {
 	    set line [string range $line 0 [expr {$i-1}]]
 	    if {$diffids ne $nullid2 && [lindex $line 1] ne "blob"} continue
 	    set sha1 [lindex $line 2]
-	    if {[string index $fname 0] eq "\""} {
-		set fname [lindex $fname 0]
-	    }
 	    lappend treeidlist($id) $sha1
 	}
+	if {[string index $fname 0] eq "\""} {
+	    set fname [lindex $fname 0]
+	}
+	set fname [encoding convertfrom $fname]
 	lappend treefilelist($id) $fname
     }
     if {![eof $gtf]} {
@@ -6296,7 +6297,7 @@ proc showfile {f} {
 	    return
 	}
     }
-    fconfigure $bf -blocking 0
+    fconfigure $bf -blocking 0 -encoding [get_path_encoding $f]
     filerun $bf [list getblobline $bf $diffids]
     $ctext config -state normal
     clear_ctext $commentend
@@ -6334,6 +6335,7 @@ proc mergediff {id} {
     global diffids
     global parents
     global diffcontext
+    global diffencoding
     global limitdiffs vfilelimit curview
 
     set diffmergeid $id
@@ -6347,9 +6349,10 @@ proc mergediff {id} {
 	error_popup "[mc "Error getting merge diffs:"] $err"
 	return
     }
-    fconfigure $mdf -blocking 0
+    fconfigure $mdf -blocking 0 -encoding binary
     set mdifffd($id) $mdf
     set np [llength $parents($curview,$id)]
+    set diffencoding [get_path_encoding {}]
     settabs $np
     filerun $mdf [list getmergediffline $mdf $id $np]
 }
@@ -6357,6 +6360,7 @@ proc mergediff {id} {
 proc getmergediffline {mdf id np} {
     global diffmergeid ctext cflist mergemax
     global difffilestart mdifffd
+    global diffencoding
 
     $ctext conf -state normal
     set nr 0
@@ -6368,18 +6372,22 @@ proc getmergediffline {mdf id np} {
 	}
 	if {[regexp {^diff --cc (.*)} $line match fname]} {
 	    # start of a new file
+	    set fname [encoding convertfrom $fname]
 	    $ctext insert end "\n"
 	    set here [$ctext index "end - 1c"]
 	    lappend difffilestart $here
 	    add_flist [list $fname]
+	    set diffencoding [get_path_encoding $fname]
 	    set l [expr {(78 - [string length $fname]) / 2}]
 	    set pad [string range "----------------------------------------" 1 $l]
 	    $ctext insert end "$pad $fname $pad\n" filesep
 	} elseif {[regexp {^@@} $line]} {
+	    set line [encoding convertfrom $diffencoding $line]
 	    $ctext insert end "$line\n" hunksep
 	} elseif {[regexp {^[0-9a-f]{40}$} $line] || [regexp {^index} $line]} {
 	    # do nothing
 	} else {
+	    set line [encoding convertfrom $diffencoding $line]
 	    # parse the prefix - one ' ', '-' or '+' for each parent
 	    set spaces {}
 	    set minuses {}
@@ -6514,7 +6522,7 @@ proc gettreediffs {ids} {
 
     set treepending $ids
     set treediff {}
-    fconfigure $gdtf -blocking 0
+    fconfigure $gdtf -blocking 0 -encoding binary
     filerun $gdtf [list gettreediffline $gdtf $ids]
 }
 
@@ -6530,6 +6538,7 @@ proc gettreediffline {gdtf ids} {
 	    if {[string index $file 0] eq "\""} {
 		set file [lindex $file 0]
 	    }
+	    set file [encoding convertfrom $file]
 	    lappend treediff $file
 	}
     }
@@ -6587,6 +6596,7 @@ proc getblobdiffs {ids} {
     global diffcontext
     global ignorespace
     global limitdiffs vfilelimit curview
+    global diffencoding
 
     set cmd [diffcmd $ids "-p -C --no-commit-id -U$diffcontext"]
     if {$ignorespace} {
@@ -6600,7 +6610,8 @@ proc getblobdiffs {ids} {
 	return
     }
     set diffinhdr 0
-    fconfigure $bdf -blocking 0
+    set diffencoding [get_path_encoding {}]
+    fconfigure $bdf -blocking 0 -encoding binary
     set blobdifffd($ids) $bdf
     filerun $bdf [list getblobdiffline $bdf $diffids]
 }
@@ -6634,6 +6645,7 @@ proc getblobdiffline {bdf ids} {
     global diffids blobdifffd ctext curdiffstart
     global diffnexthead diffnextnote difffilestart
     global diffinhdr treediffs
+    global diffencoding
 
     set nr 0
     $ctext conf -state normal
@@ -6671,10 +6683,13 @@ proc getblobdiffline {bdf ids} {
 	    } else {
 		set fname [string range $line 2 [expr {$i - 1}]]
 	    }
+	    set fname [encoding convertfrom $fname]
+	    set diffencoding [get_path_encoding $fname]
 	    makediffhdr $fname $ids
 
 	} elseif {[regexp {^@@ -([0-9]+)(,[0-9]+)? \+([0-9]+)(,[0-9]+)? @@(.*)} \
 		       $line match f1l f1c f2l f2c rest]} {
+	    set line [encoding convertfrom $diffencoding $line]
 	    $ctext insert end "$line\n" hunksep
 	    set diffinhdr 0
 
@@ -6684,6 +6699,7 @@ proc getblobdiffline {bdf ids} {
 		if {[string index $fname 0] eq "\""} {
 		    set fname [lindex $fname 0]
 		}
+		set fname [encoding convertfrom $fname]
 		set i [lsearch -exact $treediffs($ids) $fname]
 		if {$i >= 0} {
 		    setinlist difffilestart $i $curdiffstart
@@ -6694,6 +6710,8 @@ proc getblobdiffline {bdf ids} {
 		if {[string index $fname 0] eq "\""} {
 		    set fname [lindex $fname 0]
 		}
+		set fname [encoding convertfrom $fname]
+		set diffencoding [get_path_encoding $fname]
 		makediffhdr $fname $ids
 	    } elseif {[string compare -length 3 $line "---"] == 0} {
 		# do nothing
@@ -6705,6 +6723,7 @@ proc getblobdiffline {bdf ids} {
 	    $ctext insert end "$line\n" filesep
 
 	} else {
+	    set line [encoding convertfrom $diffencoding $line]
 	    set x [string range $line 0 0]
 	    if {$x == "-" || $x == "+"} {
 		set tag [expr {$x == "+"}]
-- 
1.6.0.20.g6148bc
Previous: Alexander GavrilovNext: Alexander Gavrilov
Message 3 of 8 in “Enhance encoding support.”
  1. 0/4 Enhance encoding support.Alexander Gavrilov, Oct 13, 2008
  2. 1/4 gitk: Port new encoding logic from git-gui.Alexander Gavrilov, Oct 13, 2008
  3. 2/4 gitk: Enhance file encoding support.Alexander Gavrilov, Oct 13, 2008
  4. 3/4 gitk: Implement batch lookup and caching of encoding attrs.Alexander Gavrilov, Oct 13, 2008
  5. 4/4 gitk: Optimize encoding name resolution using a lookup table.Alexander Gavrilov, Oct 13, 2008
  6. Paul MackerrasOct 15, 2008
  7. Paul MackerrasOct 15, 2008
  8. Alexander GavrilovOct 15, 2008

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.