Episode: 2011
Title: HPR2011: Introduction to sed - part 4
Source: https://hub.hackerpublicradio.org/ccdn.php?filename=/eps/hpr2011/hpr2011.mp3
Transcribed: 2025-10-18 13:18:36

---

This is HPR episode 2011 entitled Introduction to Zen Part 4 and in part on the series Learning Zen.
It is hosted by Dave Morris and in about 48 minutes long.
The summary is how Zen really works.
Let's frequently use Zen command.
This episode of HPR is brought to you by an honesthost.com.
Get 15% discount on all shared hosting with the offer code HPR15.
That's HPR15.
Better web hosting that's honest and fair at An Honesthost.com.
Hello everyone this is Dave Morris.
This is episode 4 in the series I've called an Introduction to Zen.
It's a bit more than an introduction but that's what I thought it was going to be when I started.
Anyway hope you're still with me.
So in the last episode we looked at some commands, some said commands other than the S
command that we'd spent a couple of episodes looking at and we looked at how you address lines when
you're writing a script for said.
So in this episode we're going to look at how said really works.
We've not really gone into the guts of it and we're going to have a look at some more said
commands so that we're in a position to build some useful said programs if possible.
So let's look at what said is actually doing in a bit more detail and we've done so far.
I mentioned the pattern space where said holds the incoming data in episode 1 and this is where
the data is held aligned from from whatever file it's reading.
This is where it's held while commands are being executed on it.
Remember the thing we mentioned about this.
I'm going to expand on this a little bit.
There's also another buffer or another space for holding data which is called the hold space.
So when we looked at the pattern space before I didn't really explain what it was in great detail.
We just looked at it from the point of view of it holding the current line from the input stream
but it's actually a buffer, a storage area that can hold an arbitrarily large amount of data.
It's usually holds just the input line. I have to say the way that most people use said
and but it can hold a lot more which we'll see today.
So just to clarify what is happening within said as it's processing the data that it's working on.
A line is read from the input stream. We knew that already obviously.
Then the trailing new line is removed if there is one.
And the result of that process is stored in the pattern space.
Then all of the commands making up the said script which as you know can be provided through
the command line or through files or whatever.
And these are all executed with regard to whatever the addressing specifications are.
So certain lines won't fire until you hit a certain line address and so forth.
When the command execution is all finished the pattern space is printed to the output stream
and trailing new lines added to it if one was removed.
I don't know how you test that actually. I don't know how you make the lines of that trailing new lines
but that's probably just my ignorance of Unix.
And remember that we looked at the minus n option to the command line which switches off the auto
printing that I just referred to. Then the cycle begins again after the line's been printed or not
or the new line's been grabbed and the pattern space is always cleared before a line is read
in a normal course of events but there are ways in which you can prevent that.
Now the whole space is a separate buffer, storage buffer of the same sort to the pattern space
and it's not affected by the the read and write cycle that I've just described.
So the data that you put in the whole space stays there till said exits or until you've
you delete it explicitly. There are commands that allow you to move data to and from the whole space.
So let's look at some of the commands and what I'm doing here is I'm following the
Gnu said manual which has section entitled less frequently used commands.
I'm looking at some of these. I'm not going to cover them all. Some I'll cover today
and there will be some others which will be in the next episode.
Some I won't cover at all though. Let's start with the Y command.
That's just a simple letter Y and it's followed by a delimiter which is by default is a slash,
then a sequence of characters, then a slash and another sequence of characters and a closing slash.
The first sequence of characters is called a source characters. The second sequence is called
the destination character. So what this does is it causes a said to run through the pattern space
and it transliterates or transforms characters. It looks for any that match the source character list
and it turns them into the corresponding character in the destination character list. So if it's
if the source list contains ABC and the destination contains DEF then every A will be turned into D.
When we looked at the S command a bit later on in that in that series of talking about it
we spoke about how you can change the delimiters that separate the parts of that command or
it's the same with the Y command. If you use something other than a slash as the first delimiter
then that's not a not a problem. If you actually want to have the delimiter that you're using
within either of the strings you have to proceed it with the back slash. The two lists have got to
be the same length. The source character list and the destination character list must be the same length
otherwise how could it work you know the Y command is looking for a character and when it finds one
it's looking for the corresponding character in the destination list. There are no flags for the Y command.
So I've got some examples and what I've done is to make a simple said script which
processes the first two lines of the said underscore demo 1.txt file and it processes them through
Y command and the Y command the source list is all of the lowercase vowels and the destination list
is the same vowels but shifted by one vowel. So AEIOU becomes EIOUA. So if I just read out the command
it is said space minus NE space open quote 1 comma 2 remember that means on lines 1 and 2
open curly bracket which is the start of a group of commands and we've got this Y slash AEIOU slash
EIOUA slash then we've got a semicolon and a P closed curly bracket closed quote space said
underscore demo 1.txt so what that's that's doing is it's applying the Y command to every to the
first two lines to every character on lines 1 and 2 and then it's printing the result. So what you
get is instead of hacker public radio etc I'm not going to read this out because this is really silly
is hacker pub lock rate redo and stuff like that this amazes me I'm a very small mind a childish
mind obviously but anyway it demonstrates the point I've got another one of a similar nature
it's a second example so in this case I'm using NL remember the NL command we used before which
is number lines and it numbers the lines in the same file said underscore demo 1.txt and it
feasts that said and in this particular case we have two Y commands in two groups we're using
the the one tilde two expression addressing expression meaning start on line 1 and then go two lines
after that so it will go every odd line we're doing one transformation in that case and we've got
another one starts line 2 and then goes every every second line so that'll be every even line
and we're doing a different transformation and I've got a different
delimitive for the the second Y command I'm not going to read this all out because it's really long
and I think that you should should be able to get some sort of idea from what I've just said
but there's an example that shows that it's actually working its way through the entire file doing
different transformations to every other line again it's doing this vowel twiddling business
so the next command is the equals command this is a canoe said extension and it causes said to
print out the line number the current line number from the input stream that is followed by a new
line so this is this is merely said counting the lines as they're coming in in in its cycle of
bringing in lines in and putting them in the pattern space and so on the equals command can be
proceeded by any of the address types as could the Y command just to think about that I didn't
emphasize that in the notes properly so I've got an example here using the equal command to print
out the number of the last line of the input file so it's said demo one again and the the said
expression the said command script if you like is simply dollar equals so that means do nothing
at all except on the last line print out the line number so the answer is 13 which we saw in
episode one when we were using the wc command to do the same thing just said it's perfectly
capable of doing this by itself the next example I've got prints out the line number but then it
follows it by the line and I just wanted to point out to you that the line number is always
followed by a new line so you see the the line number then you see the line and this is a said
space minus in e space then the script and the script is in quotes dollar open curly bracket
equals semicolon p close curly bracket quote so that's going to do the same thing as before except
it will print out the line as well so you see 13 on one line and detail on the topic which is
the last line of said demo one on the next line it is not straightforward to put the line number
on the front of the line within said it is possible but it's very very very convoluted I don't even
think I'm going to cover it in this series basically because there's better ways of doing that I suppose
given that we've got commands like nl so the last example in the talking about the equals command
is just pointing out that if you use the minus s option to the command you will get
and you feed it to commands feed said to two files I should say then you're going to get two ends
if you approach give it the same script so I've got said space minus s and e space then the same
script that we had before dollar curly bracket open curly bracket equals semicolon p close curly
bracket quote and we're giving it said demo one said demo two is the input files so we the answer
we get is 13 followed by detail on a topic followed by 26 followed by contribute one show a year
so it's it's operated on the two files because it's treated them separately so now we're going to
look at a group of commands that operate on the pattern space okay what we've been doing so far
also operates on the pattern space but these are some some special commands I guess you could say
that these are commands that you're not going to use all that often unless you're really into
heavy said programming so there's four commands here the first one is the D command I'm not going to
give any examples until the end here because you really need to be using them together they don't
really do very much on there on their own mostly so the D command it deletes from the pattern space
and it's related to the lowercase D this is the uppercase D I should have said it's it's related to
the lowercase D command which we've already seen but it assumes that the pattern space has multiple
new lines in it and it deletes up to the first new line then and this is where it's different the
cycle is restarted using the resulting pattern space but without reading any input so D does a
thing to the pattern space and then doesn't read more and goes back again to the start of all the
the said scripts if there is no new line in the pattern space then the whole thing is deleted
and a new cycle is begun with a new input line being read so in this case the capital D command
behaves just as the lowercase D command does and you can proceed D by any of the address types we've
already seen and we have the capital N command and this one adds the next line of input to the
pattern space preceded by a new line so it's going to append it to the pattern space if there's
no more input when the N command goes to grab input then said exits without processing any more
command so it needs some quite careful use this I I don't personally find it all that useful
I'm a bit of a said newbie I suppose so I tend not to to do stuff at this level all that often
but it's it's it's quite an advanced command then there's a capital P command and that
prints out the contents of the pattern space up to the first new line again it's assuming that
there is a pattern space that contains multiple lines separated by new lines and it prints out
the first chunk so you could do you could use the P command and the D command to sequentially print out
parts of the pattern space then there's the lowercase L command which personally I found
to be a lot more useful when you're trying to work out what a said script is doing and it prints out
what's currently in the pattern space and it does it in a special way it can be followed by a number
I've represented it as L followed by N where N is a number the pattern space is dumped out in fixed
length lines where that length is controlled by the numeric value of N so it'll print the print
amount and it will also include wherever there's a non-printable character like a new line or a
tab or something it will show them as backslash N and backslash T and so forth and the end of each
wrapped line ends with a backslash and the actual end of the line itself rather than the wrapping point
ends with a dollar and you can also control the width of the what's printed out by the L command
because there's a command line option minus lowercase L followed by a number or the alternative
version of that is minus minus line hyphen length equals and then a number and that provides a
value if N is not provided with the L command the default value for the width is 70 and if you don't
want any wrapping at all when you use this thing you can give a value of zero so you just get one
very very long line or concatenated together everything that's in the pattern space that is
so I get I have now given an example of using the L command because you can start to use that
for you you can use any of the other commands though it's not amazingly useful but I've demonstrated
it by making it print lines one and two of said demo one within a width of 80 so the script well
that let me to give you the whole line said space minus n e space open quote 1 comma 2 L
lowcase L 80 close quote and it's said demo 1 dot txt so all it really does here is to print out
those two lines with a dollar on the end there are no new lines in them because it's it's
reading one line at a time and printing them out which is merely showing the dollar at the end
now the next example is using the end command to accumulate lines into the pattern space before
dumping the result so here I've got a said command a full full command line command would be
said space minus n e space quote 1 comma 2 then open curly bracket capital N semicolon L
close curly bracket quote and said demo 1 dot txt so what you see here is the first two lines
of the file are actually printed out wrapping at 70 characters so the first line hack of a radio
blah blah blah ends with a backslash in the middle of a word and then the word continues on the
next line then you've got a backslash n at the end of the word because the line was was longer
than 70 and so it goes on until there's another line break and then the final bit of the the second
line ends with the dollar so that can be useful if you're trying to make a bit of said do stuff
and you you're struggling to understand what has accumulated in the pattern space so I couldn't
example here of using the the pattern space to construct a list it's a not official example as
most of these things are I guess the principle of what I'm doing here is I'm collecting 10 words
from the system dictionary as I refer to it user shared dict words and I'm assaging them a bit
in a bash loop then I'm feeding them to said and I'm telling it to accumulate them but only print
me the last five I could just have done five but I wouldn't mind much of an example so I've got a
full loop which iterates 10 times using a variable i and for each iteration I grab a word out of the
the dictionary using the shuff command I've done this sort of thing before another other shows
I've done and cause a lot of words in this dictionary have end in quote s so they're possessive
I just chop them off and echo the result which is the index i followed by the word so you can see
you know this is word number one two three so that's the end of the loop then I feed this just
just from a point of interest into the t command t e e the t command is it's so called because
it's like a t piece in a plumbing some of the the the water goes one way and some of it goes
the other way in the in the the t piece so in unix terms t splits the data and sends some
into a file I've got the file slash t m p slash dollar dollar and it sends the rest through
standard out into the next command in the pipeline as an aside whenever you use dollar dollar in
bash and indeed just the straight shell then dollar dollar means the current process id number
so this generates a temporary file in the temp directory which will have the number of the
the current pid process id it's a constant throughout the particular session that you're using
so it's you know you can read and write that but it's just a short hand way of making a temporary
file anyway the the after the t command is another pipe symbol which feeds the result of all this
lot into into said I haven't explained this in in minute detail I read out every character but
hopefully you'll you'll manage to to deal with that so the said command which I will read out in
detail is said space minus e space open quote capital n semicolon 1 comma 5d quote and what that
is doing is it's accumulating all 10 words into the and 10 lines perhaps a better way of putting it
into the pattern space and then for lines 1 to 5 that it's receiving it's simply deleting them so
at the end of all that lot said will auto print what's in the pattern space which will be the last
five lines of this this sequence and I made a temporary file here just so you could go and
have a look at it to see to check that the the the thing actually worked see what it was before it
was truncated let's say I made this into a file which is available as demo3.sh on the hbr site as part
of this particular show there's a link to it in the show notes to give you the full path if you
want to download it and have a look at it so the next batch of commands looking at today are
for transferring to and from the hold space I guess these coming pairs really are four of them
do anyway let's start with the lowercase h command what this does is it replaces the contents of
the hold space with the contents of the pattern space so it deletes whatever was in the hold space
beforehand and copies the pattern space contents in but the pattern space will remain as it was
you can put any of the set address types in front of this command its counterpart is the capital h
command and instead of replacing contents of the hold space it appends the pattern space to the
hold space but it precedes it by a new line it doesn't change the pattern space either
but obviously it appends to the hold space so it changes that and you can put usual address
types on front of this command the g command the lowercase g is one which transfers the hold space
into the pattern space deleting the current contents of the pattern space so it replaces the
pattern space with the contents the hold space so the hold space won't be changed but the pattern
space will be replaced and the two buffers have the same content this can be addressed in the
can have addresses on it just like the others the capital g command the counterpart of this
appends the contents the hold space to the pattern space preceded by a new line so you can see
that's the sort of corresponding thing but it's also corresponds to the current late command
then the final one in this group of five is the lowercase x command and what that simply does is it
swaps the exchanges the contents of the hold and pattern spaces so neither are destroyed there's
swapped it there exchanged so we'll start looking at some examples of how how to use these things
shortly but I wanted to talk about some of the flags and modifiers that are relevant now which I
skipped out in episodes one and two when I was talking about the s command and addresses and so
fourth in episode three one of the missing flags to the s command was capital m and there's also a
lowercase m version of it which is synonymous remember capital i and lowercase i meant the same thing
don't even know why they both exist but same here with the m capital m and lowercase m that mean
the same thing there's also a modifier which is a thing that affects the regular expression
address matching which is a capital m there isn't a lowercase m same as same as with the i as well
again the logic of this is a little strange to my mind anyway enough of that what does it mean well
it means multi-line and it's useful in the case when you're using a regular expression to match
the pattern space and it contains more than one line it's a canoe extension so when the modifier
is in place then the circumflex matches the empty string after a new line and a dollar matches
the empty string before a new line so if you've got multiple lines in your buffer then these are
matching individual lines it also allows you to use some special meta characters which
match the beginning and the end of the buffer and these are backslash back quote which is really
you're running out of characters i guess so the back quote is the one that's on the top left of
most keyboards and that's for the beginning of the buffer and backslash single quote matches the
end of the buffer so i've got an example here where i'm using these these characters to demonstrate
the point so my example is one where i've accumulated two lines in the hold space and have
transferred them to the pattern space then use an s command two s commands in fact with with g
modifiers which are not not needed in this particular example but i'm doing them anyway because
it's a will come on to that in a minute but what i'm doing is i'm putting a square bracket an
open square bracket at the beginning and a closed square bracket at the end so let me read out the
said command the command line stuff so that would be said space minus n e space quote one comma two
capital H so that means for each line in the range one to two append the line to the hold space
and remember the append to the hold space means you proceed it with a new line so after after
the capital H i've got a semi colon then a two so that means for line two do a group the group
begins with a curly open curly bracket lowercase g semi colon s slash circumflex slash open
square bracket slash g semi colon s slash dollar slash close square bracket slash g semi colon p
close curly bracket so let's summarize that what it's saying is online to do these command the lie
online to where to use the g command and you remember that the lower sorry i forget to mention
what case is lowercase g command is replace the pattern space with the contents of the hold space
so we've accumulated two lines in the hold space by the time this group is being executed
so we then grab those two lines stick them in the pattern space we then run the two s commands
which replace the beginning of the buffer with an open square bracket and the end of the buffer
with a close square bracket we haven't used the the capital M modifier here at all so we're
operating on the buffer level having done that the lowercase p command will print out whatever's
there and because there's no instructions as to what to do with the rest of the lines and we're
not in order print that's all you'll see so when you look at the output from this and this is
using said demo one again the output consists of an open square bracket on a line by itself followed
by the first line of the file followed by the second line of the file with a close square bracket
on the end of that that's because there's an extra new line at the start of the pattern space
due to the way that the h command does its thing so that just shows that in this mode of operation
there's one beginning and end in this buffer so if we do the same again but we modify the s
commands with the capital M flag flag and modifier mean the same thing really in this context but I
don't even know why they why they separate the names to be honest anyway the command is exactly
the same except that each the s commands ends with a g followed by a capital M and the result that
you get is that the first line which just contain a new line if you remember has got an open and
close square bracket the second line has got an open and square bracket on the front a close
square bracket on the end same with the third line so what this shows is that the circumflex and the
dollar relate to the new lines that are in the buffer and surround each of the lines when they're
printed out if we'd wanted to signify the start and ends of the buffer then we would need to use this
backslash back quote and backslash quote sequences meta characters this is really hard
to type on the command line because bash doesn't really deal with using the
delimiter of a string in the string very well so what I've done is I've created a file which I've
called demo for dot scd and in it I've got some commands that do the same things we've already
done but using these beginning and end of buffer characters you'll see I've in the example I've
shown what happens if you type cat demo for dot said you can see its contents it's the same
set of commands but laid out differently and when you want to invoke it you would type said
space minus n f space demo for dot said space said demo 1 dot txt and the result that you get is
the same as the first example the open square bracket occurs on the first line all by itself and
the close square bracket occurs at the end of the second line this file containing these commands
available if you want to download it and play around with it yourself and it's linked from the
the show notes so I'm going to stop talking about said commands the new ones anyway at this point
I'm not going to not going to expand on any more just now but I'm going to dive into the examples
and I've got a few examples here that try to demonstrate the use of these various commands so
example one is mainly to demonstrate how you could use the capital p command and the y command
so if I summarize the actual command that I'm using here I've got the the command line is said space
minus n e space quote open quote 1 comma 2 curly bracket open curly bracket s slash dollar slash
back slash n hyphen slash semicolon capital p semicolon y slash aio u slash eio u a slash semicolon
lowercase p close curly bracket close quote and we're using said demo 1.txt so you probably didn't
get most so hopefully you're able to look at the the notes as I'm doing this what what we're doing
here is we've got auto printing switch off with the minus n and all of the commands that we've got
here are in a group and they're controlled by an address range 1 comma 2 that just operates
on the first two lines of the file all right now the first thing we do is to use an s command
and what that does is for a given line it adds a new line at the end of it followed by a hyphen
and it's doing that in the pattern space then the capital p command so it should stop and
reiterate that by doing this we've now got in the pattern space a line the lines just been read
from the input file and we've just added a new line at the end of it with a hyphen after so
we've got effectively two lines in so the capital p command prints out the line that's some
that's just been read in it's just been edited effectively by adding a new line to the end of it
and but it only prints out up to the new line that we added so you'll see that in the output
we see the first line of the file completely untouched and so on we don't see the hyphen because
it's it's after the new line and the capital p command prints up to the new last but you say
and including but anyway I think in the but it doesn't say but I think that that when the yeah
when it says prints out up to the first new line I think it means up to ending including the
new line anyway so after that capital p command we've got a y command which is doing this this
silliness about twiddling with the vowels so the line is still in the in the pattern space we haven't
deleted it because we've not really started a cycle we haven't explicitly deleted it so it's
still there so we've just edited it with this y command and then you know as before A becomes E
E becomes i and so forth we then have the lowercase p which says print out the contents of the
pattern space which it does but the pattern space holds a the original line that was read in
with a new line after it and a hyphen after that so you then see the transformed line
followed by the hyphen and because we've then finished that group of of commands a new cycle
will begin the next line will come in and then the same thing will be done all over again the un
unchanged line will be written out then the transformation will be applied that will be printed out
with the hyphen after it so you can see again most of my examples of fairly useless but hopefully
they help to explain what these commands are actually doing the example two is demonstrating
the h and the g you capital h and the capital g command which are twiddling around with the pattern
and the whole space i'll start by reading out the command line this is quite a complex one so
you'll need to to look at itself i think to grasp it we're operating on said demo 1.txt as usual
since we do an example finally over you sorry about that anyway let's read the thing
said space minus e space quote 1 comma slash circumflex dollar slash curl open curly bracket
capital h semicolon low case d close curly bracket semicolon dollar open curly bracket
capital g semicolon s slash back slash n dollar slash slash close curly bracket close quote
and then the name of the five so we start off with an address expression that says one comma slash
circumflex dollar slash so that's every line from the first line to the first blank line that is
the one that just got a start and an end nothing else on it so the first group of commands are to
be run on those lines and the group consists of a capital h command which appends the input line
to the whole space with a new line in front of it then it's a d a low case d command which
deletes that line from the pattern space which prevents auto printing of it remember we've got
auto printing switched on we didn't have a minus n here so all of those lines are print are put
into the the cold space then all of the other lines in the file there are no there are no exact
there no addresses that operate on those particularly so all of those lines just get read in and then
printed out in the normal way that said operates okay and then the last group within the script
begins with the dollar which means do these on the last line and it consists of a capital g and
what that does is it appends the whole space to the pattern space but because when we put stuff
into the whole space we use the capital h command that causes a new line to be to be added to the
front of it so after that g command we do an s command which finds a new line backs like n
in front of a dollar which is the end of line and removes it so it's removing this the last
new line from the from the the group of lines in the in the whole space that we've pulled down
the pattern space I suppose that's actually a bit counter intuitive because what we did was to
add a add a blank line to the front of it or set added blank line to the front of it we've taken
one off the end of it well the reason we did that is because we want there to be a blank line but
we don't want it on the front we don't want it on the end we want it on the front so having done
that the we've now got pattern space with data in it we otherwise we'd have gone through the
whole file and we'd we'd be finished now we'd printed out so the effect of this has been to
stash away the first paragraph of the file in the whole space to print out the rest of the file
and then bring back the first paragraph from the whole space and print it out so what we see
in the the output is the second paragraph of the the file followed by the first paragraph file so
I hope that once clear if you're just listening to this you're trying to work or something I think
you probably had just had to struggle with this you probably even switch it off until you could look
at the notes I hope that with the notes you've actually managed to understand that this is one of
the examples that you really need to get your head around if you want to get into this level of
said I will not be upset if you say no that's enough I don't want to know any more about said by
that's fine I'm just being completist and wanting to to get to understand this myself and to
share it with you but if you don't want it then I'm perfectly happy so the final example then
the example three I haven't got a vast number of examples this time my head's about to explode I
don't like yours I'm looking here at something I mentioned in episode two where I was talking about
the solution to the problem of joining all of the lines in a text file make one long line don't
know why you'd want to do it but you know the question is could you can you well I think you can
and I think we've now got enough knowledge of said to be able to do that so let me read you my
commands is two command line commands bash commands which I should read out and then try and explain
the first one consists of a variable assignment bash variable assignment which begins x equals
this is then followed by a thing called a command substitution expression which I covered in
one of my bash tips episodes some further bash tips talked about command substitution and
the result of doing this x equals command substitution is whatever happens in the command
the result of it is to be saved in the variable x well the command is a said command would be
and it consists of said space minus n e space open quote capital H semicolon dollar open curly
bracket lowercase g semicolon s slash circumflex back slash n slash slash semicolon s slash back
slash n slash space slash g semicolon lowercase p close curly brace close quote and the file we're
operating this on is said demo one dot txt I'll come along I'll explain the said command in a moment
but we're assuming that whatever said is done to this file is now in the variable x we simply echo
that the next line is an echo command where we we give it dollar open curly brace hash mark x close
curly brace and this is a bashism which is way of reporting the length of a variable the answer is
768 so let's look at the said command then which was in this substitution it's got a minus n so
it's no water printing the h command is run on every line and this causes every input line to be
appended to the whole space where the new line in front of it you know we saw this already then
we've got a dollar and a group and that means on the last line of the file do these commands and the
the lowercase g command replaces the pattern space with the whole space okay by this time there will
be nothing in the pattern space anyway because we just did the h command on it which copied it away
to the whole space so there's nothing of any consequence so the g loc hg pulls back the the
whole space which in which are all the all the accumulated lines of the file the first s command
removes the first back slash and the first new line from the beginning of the buffer remember
that the behavior of the the cabled h commands is it precedes every line it stashes in the whole
space with the new line we don't want that one that's an extra so we're going to chop up one out
then the second s command it's got a g flag on it so it's going to do this to the whole buffer
and all it's doing is it's replacing every new line with the space when it's finished that there's
a lowercase p command which prints the whole pattern space okay so it'll return the pattern space
which now is one long line because we took all of the new lines we took the first one away through
it away and all the other ones we turned into spaces i proved in the example that well i told you
the 768 characters long the result of the first first example i described i've got another piece
of scripting here which simply does the equivalent thing using a variable y y equals dollar open bracket
cat space said underscore demo one dot txt closed parenthesis i should say it's a problem with
the way we bridge describe these brackets we call them brackets you call them parentheses if you're
in the u.s and other parts of the world anyway in dollar followed by parentheses and inside the
parentheses is the cat command so it's just copying the entirety of the file said demo one dot txt
into the y variable and if we use the echo again echo dollar open curly bracket hash y close
curly bracket we get back the number 768 which is exactly the same as we had when we run the
said thing earlier i haven't printed it out you you can uh try it out for yourself and see
what it looks like but basically it's the whole file is one long line so not all that exciting
but we solved the problem which uh i'm sure you're very happy but okay so i thought at the end of
this episode i would give you a quiz now whether you actually want to do it of course this is up to you
but um you can um send me the if you want to have a shot you can you can send me a send me the answer
as a comment or as an email or whatever the the quiz is i want you to have a go at using said
to turn text into pig Latin pig Latin is a thing from the last century i guess uh where people
used to obscure words by taking taking the first letter and sticking it on the end of putting
a a y on the end of it so pig Latin becomes igpe atinlay so the quiz is use the testator in said
demo 1.txt from episode one and convert the first line into pig Latin and the rules of how to do
this are pretty simple there should be there are exceptions in the actual rule to quite complex i
think there's more than can easily be done in said possibly done at all in said but just to have
a shot at doing them in a simplistic way so my brief rules are take the first letter of each word
place it at the end follow it with the letters a y and thus pig becomes igpe and Latin becomes
atinlay don't bother with one two and three letter words because otherwise you're going to be turning
a the letter a or word a into a a y which we don't want i suppose you could operate on three
letter words if you want to but um i'd suggest you don't don't bother about capitals ideally
Latin with a capital L should become atinlay with a capital A in the lower case L but
said's probably not the best thing to to do that with so i'll let you off that one i'll do my
solution for the next episode and uh i hope that you'll be able to come up with a much better
solution than i'm able to do because uh i'll mean yeah you're way ahead of me with the the said
all right enjoy bye
you've been listening to hecka public radio at hecka public radio dot org
we are a community podcast network that releases shows every weekday Monday through Friday
today's show like all our shows was contributed by an hbr listener like yourself
if you ever thought of recording a podcast then click on our contributing to find out
how easy it really is hecka public radio was founded by the digital dog pound and the
infonomicum computer club and it's part of the binary revolution at binrev.com if you have
comments on today's show please email the host directly leave a comment on the website or record
a follow-up episode yourself unless otherwise stated today's show is released under creative
comments attribution share a light 3.0 license