Interpretation for many more DTCs on Volvo vehicles

[aaeegg]

This adds a DTC table based on those recently provided by Richard H. Jones and Aleksi Venäläinen. The new table is a merge of the two "upstream" tables. Tips have been split off from DTC descriptions, and there is a new option to the 850 dtc command to display tips.

I processed the incoming tables as shown in the shell session below. This is less readable than @brendandburns's Python script, but it results in valid output, which the Python script doesn't. I then hand edited the result to merge and correct the lists. I placed the J1979 DTC cross-references in the tips; they seem superfluous to me, but Volvo's first-party scan tool displays such, so I left them in. I removed useless tips like "see OTP 850 DVD" - of course there's information on the OTP 850 DVD - there's information on every DTC on the OTP 850 DVD. But I left in specific references like "see OTP 850 DVD, TP-2308202, p669 and p78 and p316".

$ # Make a copy of Aleksi's table
$ cp DTC_List_850OBDII_D2.txt dtc.txt
$ # split lines that have two dtc.contains predicates
$ sed -i 's/^\(.*)\) *||* *\(dtc[^)]*\))\(.*\)/\1\3\nif (\2)\3/' dtc.txt
$ # EFI-742 has three of them, so do it again
$ sed -i 's/^\(.*)\) *||* *\(dtc[^)]*\))\(.*\)/\1\3\nif (\2)\3/' dtc.txt
$ # where dtc.contains line has an (a), (b), etc suffix, fold that into the string name
$ sed -i 's/\(-...\).*(\(.\))/\1\2/' dtc.txt
$ # adjust the string names for the multiple ecus with address 7A
$ sed -i '/EMS2000/,/####/s/EFI-/V_EFI-/' dtc.txt
$ sed -i '/M4.4/,/####/s/\(EFI-[0-9a-z]*\)/\1_7A/' dtc.txt
$ # based on the dtc.contains lines, create a space-delimited file mapping ecu name / dtc number to string name
$ # CCU and Denso don't have dtc.contains lines, so those are omitted from dtcx.txt and will be added to dtcy.txt separately
$ sed -n '/^01 - ABS/,/####/{/if/{s,.*("\([^"]*\).*//\([^-]*\)-\([0-9A-Za-z_]*\).*,abs \1 \2\3,;p}}' dtc.txt > dtcx.txt
$ sed -n '/^02 - DSA/,/####/{/if/{s,.*("\([^"]*\).*//\([^-]*\)-\([0-9A-Za-z_]*\).*,dsa \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^11 - MSA/,/####/{/if/{s,.*("\([^"]*\).*//\([^-]*\)-\([0-9A-Za-z_]*\).*,msa \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^18 - Add/,/####/{/if/{s,.*("\([^"]*\).*//\([^-]*\)-\([0-9A-Za-z_]*\).*,add \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^29 - ECC/,/####/{/if/{s,.*("\([^"]*\).*//\([^-]*\)-\([0-9A-Za-z_]*\).*,ecc \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^2D - VGLA/,/####/{/if/{s,.*("\([^"]*\).*//\([^-]*\)-\([0-9A-Za-z_]*\).*,vgla \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^2. - PS/,/####/{/if/{s,.*("\([^"]*\).*//\([^-]*\)-\([0-9A-Za-z_]*\).*,ps \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^41 - IMMO/,/####/{/if/{s,.*("\([^"]*\).*//\([^-]*\)-\([0-9A-Za-z_]*\).*,immo \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^51 - COMBI/,/####/{/if/{s,.*("\([^"]*\).*// *\([^-]*\)-\([0-9A-Za-z_]*\).*,combi \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^SRS.*96/,/^SRS.*99/{/if/{s,.*("\([^"]*\).*// *\([^-]*\)-\([0-9A-Za-z_]*\).*,srs \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^SRS.*99/,/####/{/if/{s,.*("\([^"]*\).*// *\([^-]*\)-\([0-9A-Za-z_]*\).*,srs_temic \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^ROP/,/####/{/if/{s,.*("\([^"]*\).*// *\([^-]*\)-\([0-9A-Za-z_]*\).*,rop \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^62 - RTI/,/####/{/if/{s,.*("\([^"]*\).*// *\([^-]*\)-\([0-9A-Za-z_]*\).*,rti \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^6E - Auto/,/####/{/if/{s,.*("\([^"]*\).*// *\([^-]*\)-\([0-9A-Za-z_]*\).*,aw50 \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^7A - EMS/,/####/{/if/{s,.*("\([^"]*\).*// *\([^-]*\)-\([0-9A-Za-z_]*\).*,ems2k \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ sed -n '/^7A - Mot/,/####/{/if/{s,.*("\([^"]*\).*// *\([^-]*\)-\([0-9A-Za-z_]*\).*,m44 \1 \2\3,;p}}' dtc.txt >> dtcx.txt
$ # look up the strings from dtcx.txt and generate dtcy.txt with the string names replaced by string text
$ (while read table value string; do printf %s "$table $value "; sed -n "/<string name=.$string.>/{s,.*\">\\(.*\\)</.*,\\1,;s/://;p}" dtc.txt ; done) < dtcx.txt > dtcy.txt
$ # add the CCU and Denso dtcs
$ sed -n '/^42 - CCU/,/^51 - COMBI/{/CCU-/{s/^\(..\).*\(CCU-.*\)/ccu \1 \2/; s/\.$//; p}}' dtc.txt >> dtcy.txt
$ sed -n '/B5244S2/,${/ECM-/{s/^ECM-\(....\),* \(.*\)/denso \1 ECM-\1 \2/;p}}' dtc.txt > /tmp/foo2
$ # Make a copy of Richard's table
$ cp ../export_2024-04-06_frobbed.txt dtc2.txt
$ # replace addresses with symbolic names
$ sed -i 's/^01/abs/' dtc2.txt
$ sed -i 's/^11/msa/' dtc2.txt
$ sed -i 's/^29/ecc/' dtc2.txt
$ sed -i 's/^2D/vgla/' dtc2.txt
$ sed -i 's/^2[EF]/ps/' dtc2.txt
$ sed -i 's/^41/immo/' dtc2.txt
$ sed -i 's/^51/combi/' dtc2.txt
$ sed -i 's/^58/srs/' dtc2.txt
$ sed -i 's/^6E/aw50/' dtc2.txt
$ sed -i 's/^7A/m44/' dtc2.txt
$ # remove trailing periods from descriptions, where present
$ sed -i 's/\.$//' dtc2.txt
$ # merge Richard's list with Aleksi's, retaining only one copy of duplicated entries
$ cat dtc2.txt dtcy.txt | sort | uniq > merged.txt
$ # make a list with the tips removed
$ sed 's/\\n.*//; s/\[[^]]*]//g; s/([^)]*)//g; s/  */ /g; s/ $//' merged.txt | uniq > notips.txt
$ # make a list with only the tips
$ sed -n '/\\n/{s/^\([^ ]* [^ ]* [^ ]* \)[^\\]*\\n\(.*\)/\1\2/;p }' < merged.txt > tips.txt
$ sed 's/\\n.*//' merged.txt | ruby -ne 'p=$_[/^[^ ]+ [^ ]+ [^ ]+ /]; $_.scan(/\[([^\]]*)\]/).each {|x| puts p+x[0]}' >> tips.txt
$ sed 's/\\n.*//; s/\[[^]]*]//g;' merged.txt | ruby -ne 'p=$_[/^[^ ]+ [^ ]+ [^ ]+ /]; $_.scan(/\(([^)]*)\)/).each {|x| puts p+x[0]}' >> tips.txt
$ # eliminate extra copies of duplicated tips
$ sort tips.txt | uniq > tipsuniq.txt
$ # convert the no-tips list to C source code
$ ruby -ne 'BEGIN{lastecu="";firstecu=true;};END{puts "\t{0, 0, NULL, NULL}\n};"};(ecu,hex,str,desc)=$_.chomp.split(" ",4);suf=str[/[-A-Z]+(.*)/,1];print "\t{0, 0, NULL, NULL}\n};\n\n" if !firstecu && ecu!=lastecu;firstecu=false;puts "static const struct dtc_table_entry dtc_list_#{ecu}[] = {" if ecu!=lastecu;lastecu=ecu;puts "\t{0x#{hex}, #{suf}, \"#{desc}\", NULL},"' notips.txt > foo.c

[aaeegg]

@brendandburns I believe that PR #100 represents the most complete and accurate DTC table we have at this time. Do you have any objection to this being merged instead of PR #99?

[aaeegg]

@fenugrec I'm sorry to put you in the uncomfortable position of reviewing two "competing" PRs, but as we haven't heard from @brendandburns, would you take a look at #99 and #100?

[fenugrec]

Well commits are cheap so I could start with this one which is closer to what I want to see.
The python script is certainly more elegant but IIRC it needed a bit of work. It can certainly be refined and merged later; even if it only does 90% of the job requiring a few extra sed lines, that's still an improvement vs hand-editing the whole thing. My priority here is "how will someone pull updated DTC lists from external sources, and update freediag's lists with high confidence and minimal headache".

For this PR I think it would be a good idea to preserve the steps you posted as a shell script, probably in scantool_850/ ?
Even if not really usable as-is, it'll be more useful than hidden away in a github comment that may someday be hard to find.

[aaeegg]

@fenugrec I've added the script as scantool_850/makedtc.sh.

[fenugrec]

Merged. Thanks for your patience. Will leave '99 open since a "proper" script would be a nice eventual upgrade.