Routino SVN Repository Browser

---

Revision 1444 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Mon Jul 1 16:45:48 2013 UTC (11 years, 8 months ago) by amb
File length: 15299 byte(s)
Diff to previous 1423

Re-use the Queue and Results data structure for all routes - saves a huge number
of malloc/free calls (found by valgrind/callgrind).

---

Revision 1423 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Tue Jun 25 18:17:16 2013 UTC (11 years, 8 months ago) by amb
File length: 15236 byte(s)
Diff to previous 1420

Change the ChooseSuperNodes() and MergeSuperSegments() functions to read through
the file instead of mapping the data to help slim mode (since the nodes in the
first case and segments in the second case are read sequentially).

---

Revision 1420 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Mon Jun 24 18:16:57 2013 UTC (11 years, 8 months ago) by amb
File length: 15654 byte(s)
Diff to previous 1414

Move some printf_first() messages to the front of the function and printf_last()
function to the end of the function.

---

Revision 1414 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Sat Jun 22 18:49:33 2013 UTC (11 years, 8 months ago) by amb
File length: 15654 byte(s)
Diff to previous 1410

Use SlimMapFile() and SlimUnmapFile() as the function names for the slim mode of
operation to open and close the files (to hide the fact that they are files
being opened for reading or writing unbuffered).

---

Revision 1410 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Fri Jun 21 15:48:19 2013 UTC (11 years, 8 months ago) by amb
File length: 15774 byte(s)
Diff to previous 1339

Rename the functions for unbuffered file access to make this clear.

---

Revision 1339 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Thu May 23 17:47:16 2013 UTC (11 years, 9 months ago) by amb
File length: 15614 byte(s)
Diff to previous 1297

Don't create segments when parsing input file but create the segments later
using the nodes stored in the ways file.  This makes applying changes simpler
(segments file is not kept with the --keep option) and handling changed ways is
simpler than changed segments.

---

Revision 1297 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Tue May 7 14:41:11 2013 UTC (11 years, 10 months ago) by amb
File length: 15620 byte(s)
Diff to previous 1289

Add cache functions for NodesX, SegmentsX and WaysX to speed up the
planetsplitter in slim mode.

---

Revision 1289 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Wed May 1 18:09:20 2013 UTC (11 years, 10 months ago) by amb
File length: 15255 byte(s)
Diff to previous 1287

Try to speed up the priority queue by allocating less memory and storing the
score in the queue rather than in the result.

---

Revision 1287 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Sat Apr 27 19:05:17 2013 UTC (11 years, 10 months ago) by amb
File length: 15307 byte(s)
Diff to previous 1174

Update for change to NewResultsList() function.

---

Revision 1174 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Sat Dec 1 16:17:34 2012 UTC (12 years, 3 months ago) by amb
File length: 15308 byte(s)
Diff to previous 1168

Rename the Way_* enumerated values to Highway_*, add a new Highway_None type,
change the HighwayType() function to return Highway_None instead of
Highway_Count if no match found - all changes for consistency with similar types
and functions.

---

Revision 1168 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Wed Nov 21 09:20:57 2012 UTC (12 years, 3 months ago) by amb
File length: 15304 byte(s)
Diff to previous 1166

Revert r1164 - some super-segments are longer than 65535 metres even if no
individual segment is.

---

Revision 1166 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Tue Nov 20 16:12:08 2012 UTC (12 years, 3 months ago) by amb
File length: 15365 byte(s)
Diff to previous 1164

Replace all assert statements with a custom error message that explains the
cause and suggests a solution.

---

Revision 1164 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Tue Nov 20 15:24:30 2012 UTC (12 years, 3 months ago) by amb
File length: 15201 byte(s)
Diff to previous 1161

Replace the 32-bit combined distance and flags in the segment with 16 bits for
each.

---

Revision 1161 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Tue Nov 20 14:16:58 2012 UTC (12 years, 3 months ago) by amb
File length: 15140 byte(s)
Diff to previous 1151

Tidy up all of the recent code changes - change the name of a few of the
functions.

---

Revision 1151 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Sun Nov 18 17:24:50 2012 UTC (12 years, 4 months ago) by amb
File length: 15112 byte(s)
Diff to previous 1139

Using --parse-only and --preserve must sort the data so that it is ready to
apply the changes.

---

Revision 1139 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Thu Nov 15 19:16:47 2012 UTC (12 years, 4 months ago) by amb
File length: 15120 byte(s)
Diff to previous 1123

Fixed the --preserve option.

---

Revision 1123 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Sat Nov 3 10:44:59 2012 UTC (12 years, 4 months ago) by amb
File length: 15112 byte(s)
Diff to previous 1122

Don't open the input file for appending if there is no intention to write
anything to it.

---

Revision 1122 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Sat Nov 3 08:58:47 2012 UTC (12 years, 4 months ago) by amb
File length: 15108 byte(s)
Diff to previous 1120

Change the UnmapFile() function to take a pointer to the data instead of the
filename (like the CloseFile() function takes the file descriptor).

---

Revision 1120 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Thu Nov 1 20:00:27 2012 UTC (12 years, 4 months ago) by amb
File length: 15172 byte(s)
Diff to previous 1098

Introduce a new'--append' option for appending data from a file to the currently
parsed data.  Rename the intermediate file used for storing data to be appended
to.  Add a function to call after appending to a file which closes the file and
renames it to a temporary filename which is used for the remaining processing.

---

Revision 1098 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Sat Oct 20 12:52:01 2012 UTC (12 years, 5 months ago) by amb
File length: 14900 byte(s)
Diff to previous 1089

Delete the pruned nodes before searching for super-nodes etc.

---

Revision 1089 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Wed Oct 17 18:06:33 2012 UTC (12 years, 5 months ago) by amb
File length: 14958 byte(s)
Diff to previous 1071

Rename the BitMask functions to set or clear all bits.

---

Revision 1071 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Wed Sep 19 18:58:29 2012 UTC (12 years, 6 months ago) by amb
File length: 14959 byte(s)
Diff to previous 1069

When not marking nodes that allow no traffic as super-nodes don't route through
them when calculating super-segments.

---

Revision 1069 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Mon Sep 17 17:58:47 2012 UTC (12 years, 6 months ago) by amb
File length: 14577 byte(s)
Diff to previous 1066

Do not mark barriers that cannot be passed by any type of transport as
super-nodes.

---

Revision 1066 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Thu Sep 13 18:45:56 2012 UTC (12 years, 6 months ago) by amb
File length: 14522 byte(s)
Diff to previous 1062

Update some comments and make a few very small optimisations.

---

Revision 1062 - (view) (download) (as text) (annotate) - [select for diffs]
Added Sun Sep 9 13:58:29 2012 UTC (12 years, 6 months ago) by amb
File length: 14521 byte(s)
Diff to previous 955

Tiny optimisation to super-segment calculation.