Removing first piece of a line, if there's an all cap word

30/04/2007, 23h11

Hi all,
Sorry to post a (rather) trivial question, but my brain can't find a
high(er) gear at the moment..
I have a file like:
#v+
173&&&HILLEGOM (wit bord)&
174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
175&0.1&200.6&Einde weg RE&Hyacintenlaan
176&1.3&201.9&Einde weg RE&Veenburgerweg
177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
178&1.0&203.4&ROT LI&Beeklaan, N442
179&&&DE ZILK&
180&2.0&205.4&Einde weg LI&N206
181&0.8&206.2&1ste afslag RE&Vogelaardreef
182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef
183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
184&&&NOORDWIJK&
#v-
(It's a route description for our motorclub btw)
This file is going to be imported in a laTeX file, to serve as
longtable data
Anyway, the first figure of each line is a referencei-counter, the all cap words
are place-sign names, and I would like to remove the counter on all lines with
the all-caps names, but not on other lines.
So e.g.
173&&&HILLEGOM (wit bord)&
should become:
&&&HILLEGOM (wit bord)&

I'v been over awk and ised and grep and whatnot, but I can't get something
working alas.

Thanks in advance.
Theo
--
theo at van-werkhoven.nl ICQ:277217131 SuSE Linux
linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB
"ik _heb_ niets tegen Microsoft, ik heb iets tegen
de uitwassen *van* Microsoft"

30/04/2007, 23h39

* Theo v. Werkhoven [2007.04.30 21:11]:
> So e.g.
> 173&&&HILLEGOM (wit bord)&
> should become:
> &&&HILLEGOM (wit bord)&

sed 's/^[0-9]*//' /tmp/foo

--
JR

30/04/2007, 23h41

Theo v. Werkhoven wrote:
> Hi all,
> Sorry to post a (rather) trivial question, but my brain can't find a
> high(er) gear at the moment..
> I have a file like:
> #v+
> 173&&&HILLEGOM (wit bord)&
> 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
> 175&0.1&200.6&Einde weg RE&Hyacintenlaan
> 176&1.3&201.9&Einde weg RE&Veenburgerweg
> 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
> 178&1.0&203.4&ROT LI&Beeklaan, N442
> 179&&&DE ZILK&
> 180&2.0&205.4&Einde weg LI&N206
> 181&0.8&206.2&1ste afslag RE&Vogelaardreef
> 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef
> 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
> 184&&&NOORDWIJK&
> #v-
> (It's a route description for our motorclub btw)
> This file is going to be imported in a laTeX file, to serve as
> longtable data
> Anyway, the first figure of each line is a referencei-counter, the all cap words
> are place-sign names, and I would like to remove the counter on all lines with
> the all-caps names, but not on other lines.
> So e.g.
> 173&&&HILLEGOM (wit bord)&
> should become:
> &&&HILLEGOM (wit bord)&

How is that "all cap words" defined? The fourth field (assumed & as separator)
does also contain lower case letters (within the brackets).

Here's a solution for fourth field uppercase...

awk -F\& -v OFS=\& '$4~/^[A-Z]+$/{$1=""}1'

If you just want a word with _two_ uppercase letters as criterion...

awk -F\& -v OFS=\& '$4~/^[A-Z][A-Z]/{$1=""}1'

If neither is what you want, please clarify.

Janis

>
> I'v been over awk and ised and grep and whatnot, but I can't get something
> working alas.
>
> Thanks in advance.
> Theo

01/05/2007, 05h18

Theo v. Werkhoven wrote:
> Hi all,
> Sorry to post a (rather) trivial question, but my brain can't find a
> high(er) gear at the moment..
> I have a file like:
> #v+
> 173&&&HILLEGOM (wit bord)&
> 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
> 175&0.1&200.6&Einde weg RE&Hyacintenlaan
> 176&1.3&201.9&Einde weg RE&Veenburgerweg
> 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
> 178&1.0&203.4&ROT LI&Beeklaan, N442
> 179&&&DE ZILK&
> 180&2.0&205.4&Einde weg LI&N206
> 181&0.8&206.2&1ste afslag RE&Vogelaardreef
> 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef
> 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
> 184&&&NOORDWIJK&
> #v-
> (It's a route description for our motorclub btw)
> This file is going to be imported in a laTeX file, to serve as
> longtable data
> Anyway, the first figure of each line is a referencei-counter, the all cap words
> are place-sign names, and I would like to remove the counter on all lines with
> the all-caps names, but not on other lines.
> So e.g.
> 173&&&HILLEGOM (wit bord)&
> should become:
> &&&HILLEGOM (wit bord)&
>
> I'v been over awk and ised and grep and whatnot, but I can't get something
> working alas.

$ echo "#v+
173&&&HILLEGOM (wit bord)&
174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
175&0.1&200.6&Einde weg RE&Hyacintenlaan
176&1.3&201.9&Einde weg RE&Veenburgerweg
177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
178&1.0&203.4&ROT LI&Beeklaan, N442
179&&&DE ZILK&
180&2.0&205.4&Einde weg LI&N206
181&0.8&206.2&1ste afslag RE&Vogelaardreef
182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef
183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
184&&&NOORDWIJK&
#v-
" | perl -pe's/^\d+(?=&&&[[:upper:]]+\b)//'
#v+
&&&HILLEGOM (wit bord)&
174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
175&0.1&200.6&Einde weg RE&Hyacintenlaan
176&1.3&201.9&Einde weg RE&Veenburgerweg
177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
178&1.0&203.4&ROT LI&Beeklaan, N442
&&&DE ZILK&
180&2.0&205.4&Einde weg LI&N206
181&0.8&206.2&1ste afslag RE&Vogelaardreef
182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef
183&5.4&212.3&Bij Cafe-restaurant De Witte Raaf RE ri Noordwijk 2&Duinweg
&&&NOORDWIJK&
#v-

John
--
Perl isn't a toolbox, but a small machine shop where you can special-order
certain sorts of tools at low cost and in short order. -- Larry Wall

01/05/2007, 05h50

Theo v. Werkhoven wrote:
> Hi all,
> Sorry to post a (rather) trivial question, but my brain can't find a
> high(er) gear at the moment..
> I have a file like:
> #v+
> 173&&&HILLEGOM (wit bord)&
> 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
> 175&0.1&200.6&Einde weg RE&Hyacintenlaan
> 176&1.3&201.9&Einde weg RE&Veenburgerweg
> 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
> 178&1.0&203.4&ROT LI&Beeklaan, N442
> 179&&&DE ZILK&
> 180&2.0&205.4&Einde weg LI&N206
> 181&0.8&206.2&1ste afslag RE&Vogelaardreef
> 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef
> 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
> 184&&&NOORDWIJK&
> #v-
> (It's a route description for our motorclub btw)
> This file is going to be imported in a laTeX file, to serve as
> longtable data
> Anyway, the first figure of each line is a referencei-counter, the all cap words
> are place-sign names, and I would like to remove the counter on all lines with
> the all-caps names, but not on other lines.
> So e.g.
> 173&&&HILLEGOM (wit bord)&
> should become:
> &&&HILLEGOM (wit bord)&
>
> I'v been over awk and ised and grep and whatnot, but I can't get something
> working alas.
>
> Thanks in advance.
> Theo
> --
> theo at van-werkhoven.nl ICQ:277217131 SuSE Linux
> linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB
> "ik _heb_ niets tegen Microsoft, ik heb iets tegen
> de uitwassen *van* Microsoft"

ruby -pe 'sub(/\d+(?=&&&[[:upper:]]+\b)/,"")'

01/05/2007, 06h39

* Jean-Rene David [2007.04.30 21:39]:
> * Theo v. Werkhoven [2007.04.30 21:11]:
>> So e.g.
>> 173&&&HILLEGOM (wit bord)&
>> should become:
>> &&&HILLEGOM (wit bord)&
>
> sed 's/^[0-9]*//' /tmp/foo

Sorry I misread the problem. Never mind that.

--
JR

01/05/2007, 19h20

The carbonbased lifeform John W. Krahn inspired comp.unix.shell with:
> $ echo "#v+
> 173&&&HILLEGOM (wit bord)&
> 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
> 175&0.1&200.6&Einde weg RE&Hyacintenlaan
> 176&1.3&201.9&Einde weg RE&Veenburgerweg
> 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
> 178&1.0&203.4&ROT LI&Beeklaan, N442
> 179&&&DE ZILK&
> 180&2.0&205.4&Einde weg LI&N206
> 181&0.8&206.2&1ste afslag RE&Vogelaardreef
> 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef
> 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
> 184&&&NOORDWIJK&
> #v-
> " | perl -pe's/^\d+(?=&&&[[:upper:]]+\b)//'

Thanks John, that's exactly what I had in mind.
Now that I read it it's obvious of course, but unfortunatly I'm not very
good in regexps.

Cheers,
Theo
--
theo at van-werkhoven.nl ICQ:277217131 SuSE Linux
linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB
"ik _heb_ niets tegen Microsoft, ik heb iets tegen
de uitwassen *van* Microsoft"

01/05/2007, 19h27

The carbonbased lifeform Janis Papanagnou inspired comp.unix.shell with:
> Theo v. Werkhoven wrote:
[..]
>> 173&&&HILLEGOM (wit bord)&
>> should become:
>> &&&HILLEGOM (wit bord)&
>
> How is that "all cap words" defined? The fourth field (assumed & as separator)
> does also contain lower case letters (within the brackets).

I know, I wasn't really clear. For a non-Dutch person it's not immediatly
obvious that the words following ^\d+[&&&] are names of places. Those
are the only lines I'm interested in.

> If neither is what you want, please clarify.

See John's answer.

Thanks,
Theo
--
theo at van-werkhoven.nl ICQ:277217131 SuSE Linux
linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB
"ik _heb_ niets tegen Microsoft, ik heb iets tegen
de uitwassen *van* Microsoft"

01/05/2007, 19h28

The carbonbased lifeform William James inspired comp.unix.shell with:
> ruby -pe 'sub(/\d+(?=&&&[[:upper:]]+\b)/,"")'

Tnx.
Theo
--
theo at van-werkhoven.nl ICQ:277217131 SuSE Linux
linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB
"ik _heb_ niets tegen Microsoft, ik heb iets tegen
de uitwassen *van* Microsoft"

01/05/2007, 20h29

Theo v. Werkhoven wrote:
> The carbonbased lifeform Janis Papanagnou inspired comp.unix.shell with:
>
>>Theo v. Werkhoven wrote:
>
> [..]
>
>>>173&&&HILLEGOM (wit bord)&
>>>should become:
>>>&&&HILLEGOM (wit bord)&
>>
>>How is that "all cap words" defined? The fourth field (assumed & as separator)
>>does also contain lower case letters (within the brackets).
>
>
> I know, I wasn't really clear. For a non-Dutch person it's not immediatly
> obvious that the words following ^\d+[&&&] are names of places. Those
> are the only lines I'm interested in.
>
>
>>If neither is what you want, please clarify.
>
>
> See John's answer.

Does his answer explains the requirement YOU want?

>
> Thanks,
> Theo

02/05/2007, 09h29

> Thanks John, that's exactly what I had in mind.
> Now that I read it it's obvious of course, but unfortunatly I'm not very
> good in regexps.

If you have Python, here's a more readable alternative:

for line in open("file"):
.... if "&&&" in line:
.... for ch in line:
.... if ch.isupper():
.... print line[3:].strip()
.... break
.... else:
.... print line.strip()

02/05/2007, 20h29

The carbonbased lifeform mik3l3374@gmail.com inspired comp.unix.shell with:
>
>> Thanks John, that's exactly what I had in mind.
>> Now that I read it it's obvious of course, but unfortunatly I'm not very
>> good in regexps.
>
> If you have Python, here's a more readable alternative:
>
> for line in open("file"):
> ... if "&&&" in line:
> ... for ch in line:
> ... if ch.isupper():
> ... print line[3:].strip()
> ... break
> ... else:
> ... print line.strip()

Thanks, I probably will make a Python script to do the complete
translation from csv file to LaTeX longtable file, so this comes in
handy.

Theo
--
theo at van-werkhoven.nl ICQ:277217131 SuSE Linux
linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB
"ik _heb_ niets tegen Microsoft, ik heb iets tegen
de uitwassen *van* Microsoft"

02/05/2007, 21h13

Theo v. Werkhoven wrote:
> Hi all,
> Sorry to post a (rather) trivial question, but my brain can't find a
> high(er) gear at the moment..
> I have a file like:
> #v+
> 173&&&HILLEGOM (wit bord)&
> 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
> 175&0.1&200.6&Einde weg RE&Hyacintenlaan
> 176&1.3&201.9&Einde weg RE&Veenburgerweg
> 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
> 178&1.0&203.4&ROT LI&Beeklaan, N442
> 179&&&DE ZILK&
> 180&2.0&205.4&Einde weg LI&N206
> 181&0.8&206.2&1ste afslag RE&Vogelaardreef
> 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef
> 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
> 184&&&NOORDWIJK&
> #v-
> (It's a route description for our motorclub btw)
> This file is going to be imported in a laTeX file, to serve as
> longtable data
> Anyway, the first figure of each line is a referencei-counter, the all cap words
> are place-sign names, and I would like to remove the counter on all lines with
> the all-caps names, but not on other lines.
> So e.g.
> 173&&&HILLEGOM (wit bord)&
> should become:
> &&&HILLEGOM (wit bord)&

Why is "HILLEGOM (wit bord)" in the 4th &-separated field considered
"all cap" but "ROT LI" isn't:

173&&&HILLEGOM (wit bord)&
178&1.0&203.4&ROT LI&Beeklaan, N442

> I'v been over awk and ised and grep and whatnot, but I can't get something
> working alas.

I THINK what you really want is to operate on the lines where one of the
fields is empty:

$ cat file
173&&&HILLEGOM (wit bord)&
174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
175&0.1&200.6&Einde weg RE&Hyacintenlaan
176&1.3&201.9&Einde weg RE&Veenburgerweg
177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
178&1.0&203.4&ROT LI&Beeklaan, N442
179&&&DE ZILK&
180&2.0&205.4&Einde weg LI&N206
181&0.8&206.2&1ste afslag RE&Vogelaardreef
182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven
volgen&Vogelaardreef
183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
184&&&NOORDWIJK&

$ awk '/&&/{sub(/[^&]*/,"")}1' file
&&&HILLEGOM (wit bord)&
174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
175&0.1&200.6&Einde weg RE&Hyacintenlaan
176&1.3&201.9&Einde weg RE&Veenburgerweg
177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
178&1.0&203.4&ROT LI&Beeklaan, N442
&&&DE ZILK&
180&2.0&205.4&Einde weg LI&N206
181&0.8&206.2&1ste afslag RE&Vogelaardreef
182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven
volgen&Vogelaardreef
183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
&&&NOORDWIJK&

Regards,

Ed.

02/05/2007, 21h38

The carbonbased lifeform Janis Papanagnou inspired comp.unix.shell with:
> Theo v. Werkhoven wrote:
>> The carbonbased lifeform Janis Papanagnou inspired comp.unix.shell with:
>>
>>>Theo v. Werkhoven wrote:
>>
>> [..]
>>
>>>>173&&&HILLEGOM (wit bord)&
>>>>should become:
>>>>&&&HILLEGOM (wit bord)&
>>>
>>>How is that "all cap words" defined? The fourth field (assumed & as separator)
>>>does also contain lower case letters (within the brackets).
>>
>>
>> I know, I wasn't really clear. For a non-Dutch person it's not immediatly
>> obvious that the words following ^\d+[&&&] are names of places. Those
>> are the only lines I'm interested in.
>>
>>
>>>If neither is what you want, please clarify.
>>
>>
>> See John's answer.
>
> Does his answer explains the requirement YOU want?

It does.

Theo
--
theo at van-werkhoven.nl ICQ:277217131 SuSE Linux
linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB
"ik _heb_ niets tegen Microsoft, ik heb iets tegen
de uitwassen *van* Microsoft"

02/05/2007, 22h04

The carbonbased lifeform Ed Morton inspired comp.unix.shell with:
> Theo v. Werkhoven wrote:
[..]
>> 173&&&HILLEGOM (wit bord)&
>> should become:
>> &&&HILLEGOM (wit bord)&
>
> Why is "HILLEGOM (wit bord)" in the 4th &-separated field considered
> "all cap" but "ROT LI" isn't:

You're right, they're both (all caps) of course, and your conclusion
below is what I meant.
In the result, the lines without figures between the 1st and 2nd '&',
and 2nd and 3th '&', should have the first number stripped.

> I THINK what you really want is to operate on the lines where one of the
> fields is empty:
>
> $ awk '/&&/{sub(/[^&]*/,"")}1' file
> &&&HILLEGOM (wit bord)&
> 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat
> 175&0.1&200.6&Einde weg RE&Hyacintenlaan
> 176&1.3&201.9&Einde weg RE&Veenburgerweg
> 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg
> 178&1.0&203.4&ROT LI&Beeklaan, N442
> &&&DE ZILK&
> 180&2.0&205.4&Einde weg LI&N206
> 181&0.8&206.2&1ste afslag RE&Vogelaardreef
> 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven
> volgen&Vogelaardreef
> 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg
> &&&NOORDWIJK&

It looks like your answer does what I need aswell.
Thanks.

Theo
--
theo at van-werkhoven.nl ICQ:277217131 SuSE Linux
linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB
"ik _heb_ niets tegen Microsoft, ik heb iets tegen
de uitwassen *van* Microsoft"

30/04/2007, 23h11	#1
Theo v. Werkhoven Messages: n/a Hébergeur:	Removing first piece of a line, if there's an all cap word Hi all, Sorry to post a (rather) trivial question, but my brain can't find a high(er) gear at the moment.. I have a file like: #v+ 173&&&HILLEGOM (wit bord)& 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat 175&0.1&200.6&Einde weg RE&Hyacintenlaan 176&1.3&201.9&Einde weg RE&Veenburgerweg 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg 178&1.0&203.4&ROT LI&Beeklaan, N442 179&&&DE ZILK& 180&2.0&205.4&Einde weg LI&N206 181&0.8&206.2&1ste afslag RE&Vogelaardreef 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg 184&&&NOORDWIJK& #v- (It's a route description for our motorclub btw) This file is going to be imported in a laTeX file, to serve as longtable data Anyway, the first figure of each line is a referencei-counter, the all cap words are place-sign names, and I would like to remove the counter on all lines with the all-caps names, but not on other lines. So e.g. 173&&&HILLEGOM (wit bord)& should become: &&&HILLEGOM (wit bord)& I'v been over awk and ised and grep and whatnot, but I can't get something working alas. Thanks in advance. Theo -- theo at van-werkhoven.nl ICQ:277217131 SuSE Linux linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB "ik _heb_ niets tegen Microsoft, ik heb iets tegen de uitwassen van Microsoft"

30/04/2007, 23h39	#2
Jean-Rene David Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word * Theo v. Werkhoven [2007.04.30 21:11]: > So e.g. > 173&&&HILLEGOM (wit bord)& > should become: > &&&HILLEGOM (wit bord)& sed 's/^[0-9]*//' /tmp/foo -- JR

30/04/2007, 23h41	#3
Janis Papanagnou Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word Theo v. Werkhoven wrote: > Hi all, > Sorry to post a (rather) trivial question, but my brain can't find a > high(er) gear at the moment.. > I have a file like: > #v+ > 173&&&HILLEGOM (wit bord)& > 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat > 175&0.1&200.6&Einde weg RE&Hyacintenlaan > 176&1.3&201.9&Einde weg RE&Veenburgerweg > 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg > 178&1.0&203.4&ROT LI&Beeklaan, N442 > 179&&&DE ZILK& > 180&2.0&205.4&Einde weg LI&N206 > 181&0.8&206.2&1ste afslag RE&Vogelaardreef > 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef > 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg > 184&&&NOORDWIJK& > #v- > (It's a route description for our motorclub btw) > This file is going to be imported in a laTeX file, to serve as > longtable data > Anyway, the first figure of each line is a referencei-counter, the all cap words > are place-sign names, and I would like to remove the counter on all lines with > the all-caps names, but not on other lines. > So e.g. > 173&&&HILLEGOM (wit bord)& > should become: > &&&HILLEGOM (wit bord)& How is that "all cap words" defined? The fourth field (assumed & as separator) does also contain lower case letters (within the brackets). Here's a solution for fourth field uppercase... awk -F\& -v OFS=\& '$4~/^[A-Z]+$/{$1=""}1' If you just want a word with _two_ uppercase letters as criterion... awk -F\& -v OFS=\& '$4~/^[A-Z][A-Z]/{$1=""}1' If neither is what you want, please clarify. Janis > > I'v been over awk and ised and grep and whatnot, but I can't get something > working alas. > > Thanks in advance. > Theo

01/05/2007, 05h18	#4
John W. Krahn Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word Theo v. Werkhoven wrote: > Hi all, > Sorry to post a (rather) trivial question, but my brain can't find a > high(er) gear at the moment.. > I have a file like: > #v+ > 173&&&HILLEGOM (wit bord)& > 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat > 175&0.1&200.6&Einde weg RE&Hyacintenlaan > 176&1.3&201.9&Einde weg RE&Veenburgerweg > 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg > 178&1.0&203.4&ROT LI&Beeklaan, N442 > 179&&&DE ZILK& > 180&2.0&205.4&Einde weg LI&N206 > 181&0.8&206.2&1ste afslag RE&Vogelaardreef > 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef > 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg > 184&&&NOORDWIJK& > #v- > (It's a route description for our motorclub btw) > This file is going to be imported in a laTeX file, to serve as > longtable data > Anyway, the first figure of each line is a referencei-counter, the all cap words > are place-sign names, and I would like to remove the counter on all lines with > the all-caps names, but not on other lines. > So e.g. > 173&&&HILLEGOM (wit bord)& > should become: > &&&HILLEGOM (wit bord)& > > I'v been over awk and ised and grep and whatnot, but I can't get something > working alas. $ echo "#v+ 173&&&HILLEGOM (wit bord)& 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat 175&0.1&200.6&Einde weg RE&Hyacintenlaan 176&1.3&201.9&Einde weg RE&Veenburgerweg 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg 178&1.0&203.4&ROT LI&Beeklaan, N442 179&&&DE ZILK& 180&2.0&205.4&Einde weg LI&N206 181&0.8&206.2&1ste afslag RE&Vogelaardreef 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg 184&&&NOORDWIJK& #v- " \| perl -pe's/^\d+(?=&&&[[:upper:]]+\b)//' #v+ &&&HILLEGOM (wit bord)& 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat 175&0.1&200.6&Einde weg RE&Hyacintenlaan 176&1.3&201.9&Einde weg RE&Veenburgerweg 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg 178&1.0&203.4&ROT LI&Beeklaan, N442 &&&DE ZILK& 180&2.0&205.4&Einde weg LI&N206 181&0.8&206.2&1ste afslag RE&Vogelaardreef 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef 183&5.4&212.3&Bij Cafe-restaurant De Witte Raaf RE ri Noordwijk 2&Duinweg &&&NOORDWIJK& #v- John -- Perl isn't a toolbox, but a small machine shop where you can special-order certain sorts of tools at low cost and in short order. -- Larry Wall

01/05/2007, 05h50	#5
William James Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word Theo v. Werkhoven wrote: > Hi all, > Sorry to post a (rather) trivial question, but my brain can't find a > high(er) gear at the moment.. > I have a file like: > #v+ > 173&&&HILLEGOM (wit bord)& > 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat > 175&0.1&200.6&Einde weg RE&Hyacintenlaan > 176&1.3&201.9&Einde weg RE&Veenburgerweg > 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg > 178&1.0&203.4&ROT LI&Beeklaan, N442 > 179&&&DE ZILK& > 180&2.0&205.4&Einde weg LI&N206 > 181&0.8&206.2&1ste afslag RE&Vogelaardreef > 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef > 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg > 184&&&NOORDWIJK& > #v- > (It's a route description for our motorclub btw) > This file is going to be imported in a laTeX file, to serve as > longtable data > Anyway, the first figure of each line is a referencei-counter, the all cap words > are place-sign names, and I would like to remove the counter on all lines with > the all-caps names, but not on other lines. > So e.g. > 173&&&HILLEGOM (wit bord)& > should become: > &&&HILLEGOM (wit bord)& > > I'v been over awk and ised and grep and whatnot, but I can't get something > working alas. > > Thanks in advance. > Theo > -- > theo at van-werkhoven.nl ICQ:277217131 SuSE Linux > linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB > "ik _heb_ niets tegen Microsoft, ik heb iets tegen > de uitwassen van Microsoft" ruby -pe 'sub(/\d+(?=&&&[[:upper:]]+\b)/,"")'

01/05/2007, 06h39	#6
Jean-Rene David Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word * Jean-Rene David [2007.04.30 21:39]: > * Theo v. Werkhoven [2007.04.30 21:11]: >> So e.g. >> 173&&&HILLEGOM (wit bord)& >> should become: >> &&&HILLEGOM (wit bord)& > > sed 's/^[0-9]*//' /tmp/foo Sorry I misread the problem. Never mind that. -- JR

01/05/2007, 19h20	#7
Theo v. Werkhoven Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word The carbonbased lifeform John W. Krahn inspired comp.unix.shell with: > $ echo "#v+ > 173&&&HILLEGOM (wit bord)& > 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat > 175&0.1&200.6&Einde weg RE&Hyacintenlaan > 176&1.3&201.9&Einde weg RE&Veenburgerweg > 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg > 178&1.0&203.4&ROT LI&Beeklaan, N442 > 179&&&DE ZILK& > 180&2.0&205.4&Einde weg LI&N206 > 181&0.8&206.2&1ste afslag RE&Vogelaardreef > 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef > 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg > 184&&&NOORDWIJK& > #v- > " \| perl -pe's/^\d+(?=&&&[[:upper:]]+\b)//' Thanks John, that's exactly what I had in mind. Now that I read it it's obvious of course, but unfortunatly I'm not very good in regexps. Cheers, Theo -- theo at van-werkhoven.nl ICQ:277217131 SuSE Linux linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB "ik _heb_ niets tegen Microsoft, ik heb iets tegen de uitwassen van Microsoft"

01/05/2007, 19h27	#8
Theo v. Werkhoven Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word The carbonbased lifeform Janis Papanagnou inspired comp.unix.shell with: > Theo v. Werkhoven wrote: [..] >> 173&&&HILLEGOM (wit bord)& >> should become: >> &&&HILLEGOM (wit bord)& > > How is that "all cap words" defined? The fourth field (assumed & as separator) > does also contain lower case letters (within the brackets). I know, I wasn't really clear. For a non-Dutch person it's not immediatly obvious that the words following ^\d+[&&&] are names of places. Those are the only lines I'm interested in. > If neither is what you want, please clarify. See John's answer. Thanks, Theo -- theo at van-werkhoven.nl ICQ:277217131 SuSE Linux linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB "ik _heb_ niets tegen Microsoft, ik heb iets tegen de uitwassen van Microsoft"

01/05/2007, 19h28	#9
Theo v. Werkhoven Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word The carbonbased lifeform William James inspired comp.unix.shell with: > ruby -pe 'sub(/\d+(?=&&&[[:upper:]]+\b)/,"")' Tnx. Theo -- theo at van-werkhoven.nl ICQ:277217131 SuSE Linux linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB "ik _heb_ niets tegen Microsoft, ik heb iets tegen de uitwassen van Microsoft"

01/05/2007, 20h29	#10
Janis Papanagnou Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word Theo v. Werkhoven wrote: > The carbonbased lifeform Janis Papanagnou inspired comp.unix.shell with: > >>Theo v. Werkhoven wrote: > > [..] > >>>173&&&HILLEGOM (wit bord)& >>>should become: >>>&&&HILLEGOM (wit bord)& >> >>How is that "all cap words" defined? The fourth field (assumed & as separator) >>does also contain lower case letters (within the brackets). > > > I know, I wasn't really clear. For a non-Dutch person it's not immediatly > obvious that the words following ^\d+[&&&] are names of places. Those > are the only lines I'm interested in. > > >>If neither is what you want, please clarify. > > > See John's answer. Does his answer explains the requirement YOU want? > > Thanks, > Theo

02/05/2007, 09h29	#11
mik3l3374@gmail.com Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word > Thanks John, that's exactly what I had in mind. > Now that I read it it's obvious of course, but unfortunatly I'm not very > good in regexps. If you have Python, here's a more readable alternative: for line in open("file"): .... if "&&&" in line: .... for ch in line: .... if ch.isupper(): .... print line[3:].strip() .... break .... else: .... print line.strip()

02/05/2007, 20h29	#12
Theo v. Werkhoven Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word The carbonbased lifeform mik3l3374@gmail.com inspired comp.unix.shell with: > >> Thanks John, that's exactly what I had in mind. >> Now that I read it it's obvious of course, but unfortunatly I'm not very >> good in regexps. > > If you have Python, here's a more readable alternative: > > for line in open("file"): > ... if "&&&" in line: > ... for ch in line: > ... if ch.isupper(): > ... print line[3:].strip() > ... break > ... else: > ... print line.strip() Thanks, I probably will make a Python script to do the complete translation from csv file to LaTeX longtable file, so this comes in handy. Theo -- theo at van-werkhoven.nl ICQ:277217131 SuSE Linux linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB "ik _heb_ niets tegen Microsoft, ik heb iets tegen de uitwassen van Microsoft"

02/05/2007, 21h13	#13
Ed Morton Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word Theo v. Werkhoven wrote: > Hi all, > Sorry to post a (rather) trivial question, but my brain can't find a > high(er) gear at the moment.. > I have a file like: > #v+ > 173&&&HILLEGOM (wit bord)& > 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat > 175&0.1&200.6&Einde weg RE&Hyacintenlaan > 176&1.3&201.9&Einde weg RE&Veenburgerweg > 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg > 178&1.0&203.4&ROT LI&Beeklaan, N442 > 179&&&DE ZILK& > 180&2.0&205.4&Einde weg LI&N206 > 181&0.8&206.2&1ste afslag RE&Vogelaardreef > 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef > 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg > 184&&&NOORDWIJK& > #v- > (It's a route description for our motorclub btw) > This file is going to be imported in a laTeX file, to serve as > longtable data > Anyway, the first figure of each line is a referencei-counter, the all cap words > are place-sign names, and I would like to remove the counter on all lines with > the all-caps names, but not on other lines. > So e.g. > 173&&&HILLEGOM (wit bord)& > should become: > &&&HILLEGOM (wit bord)& Why is "HILLEGOM (wit bord)" in the 4th &-separated field considered "all cap" but "ROT LI" isn't: 173&&&HILLEGOM (wit bord)& 178&1.0&203.4&ROT LI&Beeklaan, N442 > I'v been over awk and ised and grep and whatnot, but I can't get something > working alas. I THINK what you really want is to operate on the lines where one of the fields is empty: $ cat file 173&&&HILLEGOM (wit bord)& 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat 175&0.1&200.6&Einde weg RE&Hyacintenlaan 176&1.3&201.9&Einde weg RE&Veenburgerweg 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg 178&1.0&203.4&ROT LI&Beeklaan, N442 179&&&DE ZILK& 180&2.0&205.4&Einde weg LI&N206 181&0.8&206.2&1ste afslag RE&Vogelaardreef 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg 184&&&NOORDWIJK& $ awk '/&&/{sub(/[^&]*/,"")}1' file &&&HILLEGOM (wit bord)& 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat 175&0.1&200.6&Einde weg RE&Hyacintenlaan 176&1.3&201.9&Einde weg RE&Veenburgerweg 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg 178&1.0&203.4&ROT LI&Beeklaan, N442 &&&DE ZILK& 180&2.0&205.4&Einde weg LI&N206 181&0.8&206.2&1ste afslag RE&Vogelaardreef 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven volgen&Vogelaardreef 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg &&&NOORDWIJK& Regards, Ed.

02/05/2007, 21h38	#14
Theo v. Werkhoven Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word The carbonbased lifeform Janis Papanagnou inspired comp.unix.shell with: > Theo v. Werkhoven wrote: >> The carbonbased lifeform Janis Papanagnou inspired comp.unix.shell with: >> >>>Theo v. Werkhoven wrote: >> >> [..] >> >>>>173&&&HILLEGOM (wit bord)& >>>>should become: >>>>&&&HILLEGOM (wit bord)& >>> >>>How is that "all cap words" defined? The fourth field (assumed & as separator) >>>does also contain lower case letters (within the brackets). >> >> >> I know, I wasn't really clear. For a non-Dutch person it's not immediatly >> obvious that the words following ^\d+[&&&] are names of places. Those >> are the only lines I'm interested in. >> >> >>>If neither is what you want, please clarify. >> >> >> See John's answer. > > Does his answer explains the requirement YOU want? It does. Theo -- theo at van-werkhoven.nl ICQ:277217131 SuSE Linux linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB "ik _heb_ niets tegen Microsoft, ik heb iets tegen de uitwassen van Microsoft"

02/05/2007, 22h04	#15
Theo v. Werkhoven Messages: n/a Hébergeur:	Re: Removing first piece of a line, if there's an all cap word The carbonbased lifeform Ed Morton inspired comp.unix.shell with: > Theo v. Werkhoven wrote: [..] >> 173&&&HILLEGOM (wit bord)& >> should become: >> &&&HILLEGOM (wit bord)& > > Why is "HILLEGOM (wit bord)" in the 4th &-separated field considered > "all cap" but "ROT LI" isn't: You're right, they're both (all caps) of course, and your conclusion below is what I meant. In the result, the lines without figures between the 1st and 2nd '&', and 2nd and 3th '&', should have the first number stripped. > I THINK what you really want is to operate on the lines where one of the > fields is empty: > > $ awk '/&&/{sub(/[^&]/,"")}1' file > &&&HILLEGOM (wit bord)& > 174&11.1&200.5&KRS N208 Rechtdoor&Leidsestraat > 175&0.1&200.6&Einde weg RE&Hyacintenlaan > 176&1.3&201.9&Einde weg RE&Veenburgerweg > 177&0.5&202.4&1ste weg LI ri Vogelenzang&3de Loosterweg > 178&1.0&203.4&ROT LI&Beeklaan, N442 > &&&DE ZILK& > 180&2.0&205.4&Einde weg LI&N206 > 181&0.8&206.2&1ste afslag RE&Vogelaardreef > 182&0.7&206.9&VRW RE ri langevelderslag en de weg blijven > volgen&Vogelaardreef > 183&5.4&212.3&Bij Cafe-restaurant "De Witte Raaf" RE ri Noordwijk 2&Duinweg > &&&NOORDWIJK& It looks like your answer does what I need aswell. Thanks. Theo -- theo at van-werkhoven.nl ICQ:277217131 SuSE Linux linuxcounter.org: 99872 Jabber:muadib at jabber.xs4all.nl AMD XP3000+ 1024MB "ik _heb_ niets tegen Microsoft, ik heb iets tegen de uitwassen van* Microsoft"

Outils de la discussion
Afficher une version imprimable Envoyer un lien vers cette page par email