Please check if help is okay
Hi all, in r3963 I tried to document the new prefix / suffix stuff: --road-name-config=file This option handles the problem that some countries have road names which ofter start or end with very similar words, e.g. in France the first word is very often 'Rue', ofter followed by a preposition like 'de la' or 'des'. This leads to rather long road names like 'Rue de la Concorde' where only the word 'Concorde' is really interesting. In the USA, you ofter have names like 'West Main Street' where only the word 'Main' is important. Garmin software has some tricks to handle this problem. It allows to use special characters in the road labels which mark the beginning and end of the important part. There are two different visiual effects of this option: - On the PC, when zooming out, the name 'Rue de la Concorde' is only rendered as 'Concorde'. - The index for road names only contains the important part of the name. You can search for road name Conc to find road names like 'Rue de la Concorde'. One problem: Search for 'Rue' will not list 'Rue de la Concorde' or 'Rue du Moulin'. It may list 'Rueben Brookins Road' if that is in the map. Another effect is that the index is smaller. The option specifies the path to a file which gives the details. See comments in the sample roadNameConfig.txt for further details. Is that clear enough? Please check also the comments in the sample file. Please suggest improvements, else I'll use the same text for the style documentation before merging to trunk. Gerd
Hi Gerd, it is clear your documentation for the new --road-name-config parameter. However, what is the relation or impact using this new parameter concomitantly with --x-split-name-index? Are they mutually exclusive? I suggest to clarify this in the documentation. Thanks. Regards, Alexandre 2017-05-31 4:39 GMT-03:00 Gerd Petermann <GPetermann_muenchen@hotmail.com>:
Hi all,
in r3963 I tried to document the new prefix / suffix stuff: --road-name-config=file This option handles the problem that some countries have road names which ofter start or end with very similar words, e.g. in France the first word is very often 'Rue', ofter followed by a preposition like 'de la' or 'des'. This leads to rather long road names like 'Rue de la Concorde' where only the word 'Concorde' is really interesting. In the USA, you ofter have names like 'West Main Street' where only the word 'Main' is important. Garmin software has some tricks to handle this problem. It allows to use special characters in the road labels which mark the beginning and end of the important part. There are two different visiual effects of this option: - On the PC, when zooming out, the name 'Rue de la Concorde' is only rendered as 'Concorde'. - The index for road names only contains the important part of the name. You can search for road name Conc to find road names like 'Rue de la Concorde'. One problem: Search for 'Rue' will not list 'Rue de la Concorde' or 'Rue du Moulin'. It may list 'Rueben Brookins Road' if that is in the map.
Another effect is that the index is smaller.
The option specifies the path to a file which gives the details. See comments in the sample roadNameConfig.txt for further details.
Is that clear enough? Please check also the comments in the sample file. Please suggest improvements, else I'll use the same text for the style documentation before merging to trunk.
Gerd _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Alexandre, thanks for the hint. The options work fine together. I'll try to make that clearer. Gerd ________________________________________ Von: mkgmap-dev <mkgmap-dev-bounces@lists.mkgmap.org.uk> im Auftrag von Alexandre Loss <alexandre.loss@gmail.com> Gesendet: Mittwoch, 31. Mai 2017 18:06:24 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Please check if help is okay Hi Gerd, it is clear your documentation for the new --road-name-config parameter. However, what is the relation or impact using this new parameter concomitantly with --x-split-name-index? Are they mutually exclusive? I suggest to clarify this in the documentation. Thanks. Regards, Alexandre 2017-05-31 4:39 GMT-03:00 Gerd Petermann <GPetermann_muenchen@hotmail.com<mailto:GPetermann_muenchen@hotmail.com>>: Hi all, in r3963 I tried to document the new prefix / suffix stuff: --road-name-config=file This option handles the problem that some countries have road names which ofter start or end with very similar words, e.g. in France the first word is very often 'Rue', ofter followed by a preposition like 'de la' or 'des'. This leads to rather long road names like 'Rue de la Concorde' where only the word 'Concorde' is really interesting. In the USA, you ofter have names like 'West Main Street' where only the word 'Main' is important. Garmin software has some tricks to handle this problem. It allows to use special characters in the road labels which mark the beginning and end of the important part. There are two different visiual effects of this option: - On the PC, when zooming out, the name 'Rue de la Concorde' is only rendered as 'Concorde'. - The index for road names only contains the important part of the name. You can search for road name Conc to find road names like 'Rue de la Concorde'. One problem: Search for 'Rue' will not list 'Rue de la Concorde' or 'Rue du Moulin'. It may list 'Rueben Brookins Road' if that is in the map. Another effect is that the index is smaller. The option specifies the path to a file which gives the details. See comments in the sample roadNameConfig.txt for further details. Is that clear enough? Please check also the comments in the sample file. Please suggest improvements, else I'll use the same text for the style documentation before merging to trunk. Gerd _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk<mailto:mkgmap-dev@lists.mkgmap.org.uk> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi all, okay, I've now also added options --mdr7-del, --mdr7-excl, and --poi-excl-index. Option --split-name-index is no longer temporay. The corresponding part of the help now looks like this: --split-name-index An option to enable indexing each part of a street name separately. So for example if the street is "Aleksandra Gryglewskiego" then you will be able to search for it as both "Aleksandra" and "Gryglewskiego". It will also increase the size of the index. Useful in countries where searching for the first word in name is not the right thing to do. Words following an opening bracket '(' are ignored. See also option road-name-config. --road-name-config=file This option handles the problem that some countries have road names which ofter start or end with very similar words, e.g. in France the first word is very often 'Rue', ofter followed by a preposition like 'de la' or 'des'. This leads to rather long road names like 'Rue de la Concorde' where only the word 'Concorde' is really interesting. In the USA, you ofter have names like 'West Main Street' where only the word 'Main' is important. Garmin software has some tricks to handle this problem. It allows to use special characters in the road labels which mark the beginning and end of the important part. In combinarion with option split-name-index only the words in the important part are indexed. There are two different visiual effects of this option: - On the PC, when zooming out, the name 'Rue de la Concorde' is only rendered as 'Concorde'. - The index for road names only contains the important part of the name. You can search for road name Conc to find road names like 'Rue de la Concorde'. One problem: Search for 'Rue' will not list 'Rue de la Concorde' or 'Rue du Moulin'. It may list 'Rueben Brookins Road' if that is in the map. Only MapSource shows a corresponding hint. Another effect is that the index is smaller. The option specifies the path to a file which gives the details. See comments in the sample roadNameConfig.txt for further details. --mdr7-excl This option allows to specify words which should not be in the road index. It was added before option road-name-config and is probably no longer needed. Example usage: --x-mdr7-excl="Road, Street, Straße, Weg" --mdr7-del Use this option if your style adds strings to the labels of roads which you want to see in the map but which should not appear in the result list of a road name / address search. The list is used like this: For each road label mkgmap searches the last blank. If one is found, it checks if the word after it appears in the given list. If so, the word is removed and the search is repeated. The remaining string is used to create the index. --poi-excl-index By default, mkgmap indexes the following POI types with a non-empty label: - 0x00 .. 0x0f (cities, sub type 0, type <= 0xf) - 0x2axx..0x30xx (Food & Drink, Lodging, ...) - 0x28xx (no category ?) - 0x64xx .. 0x66xx (attractions) This option allows to exclude POI types from the index. The exclueded types are not indexed, but may still be searchable on a device as some devices seem to ignore most of the index, e.g. an Oregon 600 with firmware 5.00 only seems to use it for city search. If you device finds a POI name like 'Planet' when you search for 'Net' it doesn't use the index because the index created by mkgmap cannot help for that search. So, this option may help when you care about the size of the index or the memory that is needed to calculate it. The option expects a comma separated list of types or type ranges. A range is given with from-type-to-type, e.g. 0x6400-0x6405. First and last type are both excluded.A range can span multiple types, e.g. 0x6400-0x661f. Examples for usage: - Assume your style adds a POI with type 0x2800 for each addr:housenumber. It is not useful to index those numbers, so you can use --poi-excl-index=0x2800 to exclude this. - For the mentioned Oregon you may use --poi-excl-index=0x2a00-0x661f to reduce the index size. Please suggest improvements. ciao, Gerd ________________________________________ Von: mkgmap-dev <mkgmap-dev-bounces@lists.mkgmap.org.uk> im Auftrag von Gerd Petermann <GPetermann_muenchen@hotmail.com> Gesendet: Mittwoch, 31. Mai 2017 18:35:36 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Please check if help is okay Hi Alexandre, thanks for the hint. The options work fine together. I'll try to make that clearer. Gerd ________________________________________ Von: mkgmap-dev <mkgmap-dev-bounces@lists.mkgmap.org.uk> im Auftrag von Alexandre Loss <alexandre.loss@gmail.com> Gesendet: Mittwoch, 31. Mai 2017 18:06:24 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Please check if help is okay Hi Gerd, it is clear your documentation for the new --road-name-config parameter. However, what is the relation or impact using this new parameter concomitantly with --x-split-name-index? Are they mutually exclusive? I suggest to clarify this in the documentation. Thanks. Regards, Alexandre 2017-05-31 4:39 GMT-03:00 Gerd Petermann <GPetermann_muenchen@hotmail.com<mailto:GPetermann_muenchen@hotmail.com>>: Hi all, in r3963 I tried to document the new prefix / suffix stuff: --road-name-config=file This option handles the problem that some countries have road names which ofter start or end with very similar words, e.g. in France the first word is very often 'Rue', ofter followed by a preposition like 'de la' or 'des'. This leads to rather long road names like 'Rue de la Concorde' where only the word 'Concorde' is really interesting. In the USA, you ofter have names like 'West Main Street' where only the word 'Main' is important. Garmin software has some tricks to handle this problem. It allows to use special characters in the road labels which mark the beginning and end of the important part. There are two different visiual effects of this option: - On the PC, when zooming out, the name 'Rue de la Concorde' is only rendered as 'Concorde'. - The index for road names only contains the important part of the name. You can search for road name Conc to find road names like 'Rue de la Concorde'. One problem: Search for 'Rue' will not list 'Rue de la Concorde' or 'Rue du Moulin'. It may list 'Rueben Brookins Road' if that is in the map. Another effect is that the index is smaller. The option specifies the path to a file which gives the details. See comments in the sample roadNameConfig.txt for further details. Is that clear enough? Please check also the comments in the sample file. Please suggest improvements, else I'll use the same text for the style documentation before merging to trunk. Gerd _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk<mailto:mkgmap-dev@lists.mkgmap.org.uk> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Perfect now! Thanks! Alexandre Enviado do meu iPhone
Em 1 de jun de 2017, às 03:38, Gerd Petermann <GPetermann_muenchen@hotmail.com> escreveu:
Hi all,
okay, I've now also added options --mdr7-del, --mdr7-excl, and --poi-excl-index. Option --split-name-index is no longer temporay. The corresponding part of the help now looks like this:
--split-name-index An option to enable indexing each part of a street name separately. So for example if the street is "Aleksandra Gryglewskiego" then you will be able to search for it as both "Aleksandra" and "Gryglewskiego". It will also increase the size of the index. Useful in countries where searching for the first word in name is not the right thing to do. Words following an opening bracket '(' are ignored. See also option road-name-config.
--road-name-config=file This option handles the problem that some countries have road names which ofter start or end with very similar words, e.g. in France the first word is very often 'Rue', ofter followed by a preposition like 'de la' or 'des'. This leads to rather long road names like 'Rue de la Concorde' where only the word 'Concorde' is really interesting. In the USA, you ofter have names like 'West Main Street' where only the word 'Main' is important. Garmin software has some tricks to handle this problem. It allows to use special characters in the road labels which mark the beginning and end of the important part. In combinarion with option split-name-index only the words in the important part are indexed.
There are two different visiual effects of this option: - On the PC, when zooming out, the name 'Rue de la Concorde' is only rendered as 'Concorde'. - The index for road names only contains the important part of the name. You can search for road name Conc to find road names like 'Rue de la Concorde'. One problem: Search for 'Rue' will not list 'Rue de la Concorde' or 'Rue du Moulin'. It may list 'Rueben Brookins Road' if that is in the map. Only MapSource shows a corresponding hint.
Another effect is that the index is smaller. The option specifies the path to a file which gives the details. See comments in the sample roadNameConfig.txt for further details.
--mdr7-excl This option allows to specify words which should not be in the road index. It was added before option road-name-config and is probably no longer needed. Example usage: --x-mdr7-excl="Road, Street, Straße, Weg"
--mdr7-del Use this option if your style adds strings to the labels of roads which you want to see in the map but which should not appear in the result list of a road name / address search. The list is used like this: For each road label mkgmap searches the last blank. If one is found, it checks if the word after it appears in the given list. If so, the word is removed and the search is repeated. The remaining string is used to create the index.
--poi-excl-index By default, mkgmap indexes the following POI types with a non-empty label: - 0x00 .. 0x0f (cities, sub type 0, type <= 0xf) - 0x2axx..0x30xx (Food & Drink, Lodging, ...) - 0x28xx (no category ?) - 0x64xx .. 0x66xx (attractions) This option allows to exclude POI types from the index. The exclueded types are not indexed, but may still be searchable on a device as some devices seem to ignore most of the index, e.g. an Oregon 600 with firmware 5.00 only seems to use it for city search. If you device finds a POI name like 'Planet' when you search for 'Net' it doesn't use the index because the index created by mkgmap cannot help for that search. So, this option may help when you care about the size of the index or the memory that is needed to calculate it. The option expects a comma separated list of types or type ranges. A range is given with from-type-to-type, e.g. 0x6400-0x6405. First and last type are both excluded.A range can span multiple types, e.g. 0x6400-0x661f. Examples for usage: - Assume your style adds a POI with type 0x2800 for each addr:housenumber. It is not useful to index those numbers, so you can use --poi-excl-index=0x2800 to exclude this. - For the mentioned Oregon you may use --poi-excl-index=0x2a00-0x661f to reduce the index size.
Please suggest improvements.
ciao, Gerd
________________________________________ Von: mkgmap-dev <mkgmap-dev-bounces@lists.mkgmap.org.uk> im Auftrag von Gerd Petermann <GPetermann_muenchen@hotmail.com> Gesendet: Mittwoch, 31. Mai 2017 18:35:36 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Please check if help is okay
Hi Alexandre,
thanks for the hint. The options work fine together. I'll try to make that clearer.
Gerd ________________________________________ Von: mkgmap-dev <mkgmap-dev-bounces@lists.mkgmap.org.uk> im Auftrag von Alexandre Loss <alexandre.loss@gmail.com> Gesendet: Mittwoch, 31. Mai 2017 18:06:24 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Please check if help is okay
Hi Gerd,
it is clear your documentation for the new --road-name-config parameter. However, what is the relation or impact using this new parameter concomitantly with --x-split-name-index? Are they mutually exclusive?
I suggest to clarify this in the documentation.
Thanks.
Regards,
Alexandre
2017-05-31 4:39 GMT-03:00 Gerd Petermann <GPetermann_muenchen@hotmail.com<mailto:GPetermann_muenchen@hotmail.com>>: Hi all,
in r3963 I tried to document the new prefix / suffix stuff: --road-name-config=file This option handles the problem that some countries have road names which ofter start or end with very similar words, e.g. in France the first word is very often 'Rue', ofter followed by a preposition like 'de la' or 'des'. This leads to rather long road names like 'Rue de la Concorde' where only the word 'Concorde' is really interesting. In the USA, you ofter have names like 'West Main Street' where only the word 'Main' is important. Garmin software has some tricks to handle this problem. It allows to use special characters in the road labels which mark the beginning and end of the important part. There are two different visiual effects of this option: - On the PC, when zooming out, the name 'Rue de la Concorde' is only rendered as 'Concorde'. - The index for road names only contains the important part of the name. You can search for road name Conc to find road names like 'Rue de la Concorde'. One problem: Search for 'Rue' will not list 'Rue de la Concorde' or 'Rue du Moulin'. It may list 'Rueben Brookins Road' if that is in the map.
Another effect is that the index is smaller.
The option specifies the path to a file which gives the details. See comments in the sample roadNameConfig.txt for further details.
Is that clear enough? Please check also the comments in the sample file. Please suggest improvements, else I'll use the same text for the style documentation before merging to trunk.
Gerd _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk<mailto:mkgmap-dev@lists.mkgmap.org.uk> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
participants (2)
-
Alexandre Loss -
Gerd Petermann