Suppose I have an output which is a large array, say 500x4.
I would like to transform this array into code form to create the same array.
An example:
output_array = randn(2,3)
output_array =
-0.2408 -0.9390 1.1794
-0.9064 0.1178 -0.4643
I would like to obtain:
input_array = [-0.2408 -0.9390 1.1794; -0.9064 0.1178 -0.4643]
For low dimensions, manual work is easy, but I would to be able to obtain the input code for arbitrary matrices.
Thanks.
This is one way:
a = randi(2,3)
linespec = [repmat('%d ',1,size(a,2)) '; ']
spec = strrep(['[ ' repmat(linespec,1,size(a,1)) ' ]'],'; ','')
b = sprintf(spec,a(:) )
a =
2 2 1
1 1 1
1 2 2
linespec =
%d %d %d ;
spec =
[ %d %d %d ; %d %d %d ; %d %d %d ]
b =
[ 2 1 1 ; 2 1 2 ; 1 1 2 ]
Be aware, that the number of spaces defined in ' ... ' snippets do matter!
use:
spec = ['b = ' strrep(['[ ' repmat(linespec,1,size(a,1)) ' ]'],'; ','') ]
to get:
b =
b = [ 1 2 2 ; 1 2 2 ; 1 2 1 ]
It could be interesting to wrap everything into a function:
function [ out ] = Matrix2String( a )
varname = inputname(1);
linespec = [repmat('%d ',1,size(a,2)) '; '];
spec = [varname ' = ' strrep(['[ ' repmat(linespec,1,size(a,1)) ' ]'],'; ','') ];
out = sprintf(spec,a(:) );
end
a call like:
a = randi(2,3)
Matrix2String(a)
would then return:
ans =
a = [ 2 1 1 ; 2 1 2 ; 1 1 1 ]
Related
I took a QB45 application and produced a QB64 application from it over 100,000 lines of code. All the pricing was hard coded in the program so I started reading from CSV files but now I need the date from the CSV file without putting a directory to a text file and reading for the date. I found this code below
The problem when I run it is that the assembler is for 16 bit registers and I am using 32 or 64 with windows 7 and i5 core. Can anyone help me to figure out how the date will be returned in a longer int value from the register?
'===========================================================================
' Subject: GET/SET FILE DATE/TIME Date: Unknown Date (00:00)
' Author: Matt Hart Code: QB, PDS
' Keys: GET,SET,FILE,DATE,TIME Packet: DOS.ABC
'===========================================================================
' FILEDATE.BAS by Matt Hart
'
' Gets or sets a file date/time
'
' GetFileDateTime returns the Date in MM-DD-YYYY format
' and the Time in HH:MM:SS
' SetFileDateTime expects the Date and Time in the same formats
'$INCLUDE: 'QB.BI' ' Use your path to QB or QBX.BI
DEFINT A-Z
DECLARE SUB GetFileDateTime (F$, Dat$, Tim$, Ecode%)
DECLARE SUB SetFileDateTime (F$, Dat$, Tim$, Ecode%)
' ------------------------- Sample code
F$ = LTRIM$(RTRIM$(COMMAND$))
CALL GetFileDateTime(F$, Dat$, Tim$, Ecode)
IF NOT Ecode THEN
PRINT F$; " date is "; Dat$
PRINT F$; " time is "; Tim$
ELSE
PRINT "1 Error = "; Ecode
END
END IF
NewTim$ = "01:01:02"
NewDat$ = "02-02-1980"
CALL SetFileDateTime(F$, NewDat$, NewTim$, Ecode)
IF Ecode THEN
PRINT "2 Error = "; Ecode
END
END IF
CALL GetFileDateTime(F$, Dat$, Tim$, Ecode)
IF Ecode THEN
PRINT "3 Error = "; Ecode
END
END IF
PRINT F$; " new date is "; Dat$
PRINT F$; " new time is "; Tim$
CALL SetFileDateTime(F$, Dat$, Tim$, Ecode)
IF Ecode THEN
PRINT "4 Error = "; Ecode
END
END IF
END
' ------------------------------------
SUB GetFileDateTime (F$, Dat$, Tim$, Ecode)
Ecode = 0
DIM InRegs AS RegTypeX
DIM OutRegs AS RegTypeX
InRegs.ax = &H3D00 ' Open file function
DIM FileName AS STRING * 128 ' Use fixed length
FileName = F$ + CHR$(0) ' Must be ASCIIZ string
InRegs.ds = VARSEG(FileName) ' Fixed length makes these
InRegs.dx = VARPTR(FileName) ' come out right
CALL INTERRUPTX(&H21, InRegs, OutRegs) ' Open the file
IF NOT OutRegs.flags THEN ' No error
Handle = OutRegs.ax ' Save DOS file handle
InRegs.ax = &H5700 ' Get date/time function
InRegs.bx = Handle
CALL INTERRUPTX(&H21, InRegs, OutRegs)
HMS& = OutRegs.cx ' Use long integer for
IF HMS& < 0& THEN HMS& = 65536 + HMS& ' positive numbers
Hours = HMS& \ 2048& ' Hours is first 5 bits
Minutes = (HMS& AND 2047&) \ 31& ' Minutes is next 6 bits
Seconds = HMS& AND 31& ' Seconds is last 5 bits
H$ = LTRIM$(STR$(Hours))
M$ = LTRIM$(STR$(Minutes)): IF LEN(M$) = 1 THEN M$ = "0" + M$
S$ = LTRIM$(STR$(Seconds)): IF LEN(S$) = 1 THEN S$ = "0" + S$
Tim$ = H$ + ":" + M$ + ":" + S$
YMD& = OutRegs.dx ' Long int here too
IF YMD& < 0 THEN YMD& = 65536 + YMD& ' Convert to + if needed
Year = 1980& + YMD& \ 512& ' Year is first 7 bits
Month = (YMD& AND 511&) \ 31& ' Month is next 4 bits
Day = YMD& AND 31& ' Day is last 5 bits
Y$ = LTRIM$(STR$(Year))
M$ = LTRIM$(STR$(Month))
D$ = LTRIM$(STR$(Day)): IF LEN(D$) = 1 THEN D$ = "0" + D$
Dat$ = M$ + "-" + D$ + "-" + Y$
InRegs.ax = &H3E00 ' Close file function
InRegs.bx = Handle
CALL INTERRUPTX(&H21, InRegs, OutRegs) ' Close it
ELSE
Ecode = OutRegs.flags ' Otherwise return error flags
END IF
END SUB
SUB SetFileDateTime (F$, Dat$, Tim$, Ecode)
Ecode = 0
DIM InRegs AS RegTypeX
DIM OutRegs AS RegTypeX
InRegs.ax = &H3D00
DIM FileName AS STRING * 128
FileName = F$ + CHR$(0)
InRegs.ds = VARSEG(FileName)
InRegs.dx = VARPTR(FileName)
CALL INTERRUPTX(&H21, InRegs, OutRegs)
IF NOT OutRegs.flags THEN
Handle = OutRegs.ax
InRegs.ax = &H5701
InRegs.bx = Handle
Hours& = VAL(LEFT$(Tim$, 2)) * 2048&
Minutes& = VAL(MID$(Tim$, 4, 2)) * 32&
Seconds& = VAL(RIGHT$(Tim$, 2)) \ 2
HMS& = Hours& + Minutes& + Seconds&
IF HMS& > 65536 THEN
InRegs.cx = 65536 - HMS&
ELSE
InRegs.cx = HMS&
END IF
Year& = (VAL(RIGHT$(Dat$, 4)) - 1980&) * 512&
Month& = VAL(LEFT$(Dat$, 2)) * 32&
Day& = VAL(MID$(Dat$, 4, 2))
YMD& = Year& + Month& + Day&
IF YMD& > 65536 THEN
InRegs.dx = 65536 - YMD&
ELSE
InRegs.dx = YMD&
END IF
CALL INTERRUPTX(&H21, InRegs, OutRegs)
InRegs.ax = &H3E00
InRegs.bx = Handle
CALL INTERRUPTX(&H21, InRegs, OutRegs)
ELSE
Ecode = OutRegs.flags
END IF
END SUB
I don't know why but I am getting this error:
Error in mr_lsbpex (line 3)
dlen = uint32(0) ;
Output argument "a" (and maybe others) not assigned during call to "E:\path\mr_lsbpex.m>mr_lsbpex"
I have tested "dlen = uint32(0) ;" in matlab enviorment (outside of this function) and everything was OK. Here is my code:
function a = mr_lsbpex ( r, p )
% extract from an array
dlen = uint32(0) ;
s = size (r) ;
rnd = rand (s(1),s(2)) ;
rd = 32 ;
rl = s(2) ;
for i=1:s(2)
if rnd(1,i)<rd/rl
d = bitget (round(r(1,i)/p),1);
dlen = bitset (dlen,rd,d);
rd = rd -1 ;
end
rl = rl -1 ;
end
if (dlen > 10000000 )
clear a ;
return ;
end
a = uint8(zeros(dlen,1)) ;
rd = double(dlen * 8) ;
rl = double(s(1)*s(2)-s(2)) ;
for i=2:s(1)
for j=1:s(2)
if rnd(i,j)<rd/rl
d = bitget (round(r(i,j)/p) ,1) ;
a = z_set_bit (a,rd,d) ;
rd = rd - 1 ;
end
rl = rl - 1 ;
end
end
Remember: a needs to be returned ALLWAYS!
The error is not in that specific line, but in the "whole" function itself.
Your problem is that Matlab thinks that a its not going to be created. And actually in some case it may not be created.
The following line in the beginning of your function should do the trick
a=0; % well, or a=NaN; or whatever you want to return
Additionally, don't clear a in if (dlen > 10000000 ).
I want to make the results like this:
phraseblanks
phrasemat = Hello and how are you?
Hi there everyone!
How is it going?
WHazzup?
Phrase 1 had 4 blanks
Phrase 2 had 3 blanks
Phrase 3 had 2 blanks
Phrase 4 had 0 blanks
New phrasemat is :
Hello&and&how&are&you?
Hi&there&everyone!
How&is&it&going?
WHazzup?
so I made script "phraseblanks.m":
phrasemat = char('Hello and how are you?', ...
'Hi there everyone!', 'How is it going?', 'WHazzup?')
[r, c] = size(phrasemat);
for i = 1:r
phrasemat_new = cell(r, c);
howmany = countblanks(phrasemat(i, :));
fprintf('Phrase %d had %d blanks\n', i, howmany);
phrasemat(i,:) = strrep(phrasemat(i,:),' ','&')
phrasemat_new{i,:} = [phrasemat(i,:)];
end
fprintf('Changing one is %s\n', eval('phrasemat_new'));
script "countblanks.m":
function num = countblanks(phrase)
% countblanks returns the # of blanks in a trimmed string
% Format: countblanks(string)
num = length(strfind(strtrim(phrase), ' '));
end
and I keep having errors.
please help me..
I modified slightly your phraseblanks.m so that it works.
phrasemat = {'Hello and how are you?', ...
'Hi there everyone!', ...
'How is it going?', ...
'WHazzup?'};
r = numel(phrasemat);
phrasemat_new = cell(1, r);
for i = 1:r
howmany = countblanks(phrasemat{i});
fprintf('Phrase %d had %d blanks\n', i, howmany);
phrasemat{i} = strrep(phrasemat{i}, ' ', '&');
phrasemat_new(i) = phrasemat(i);
end
fprintf('Changing one is %s\n', phrasemat_new{:});
Obviously it could be written in a nicer, more "matlaby-way", but I didn't want to stride too far from your original version. Also you could consider using regular expressions, since if you have to spaces next to each other and you want to treat them as one blank space.
I have a CSV file in a format similar to the following one:
title1
index columnA1 columnA2 columnA3
1 2 3 6
2 23 23 1
3 2 3 45
4 2 2 101
title2
index columnB1 columnB2 columnB3
1 23 53 6
2 22 13 1
3 5 4 43
4 8 6 102
I want to build a function readCustomCSV which receives a CSV file in the bellow illustrated format and a row index i and returns an output file with (for let's say i = 3) the following content:
title1
index columnA1 columnA2 columnA3
3 2 3 45
title2
index columnB1 columnB2 columnB3
3 5 4 43
Do you know how to use the csvread function in order to obtain this type of functionality?
It confuses me that there are 2 types sections. I was thinking at using the whole thing as a string and then split it into 2 .csv files and then read the corresponding line line.
try using this function :
I assumed that all tables have equal number of columns/rows. The code can definitely be shortened / improved / extended ;)
function multi_table_csvread (row_index)
filename_INPUT = 'multi_table.csv' ;
filename_OUTPUT = 'selected_row.csv' ;
fIN = fopen(filename_INPUT,'r');
nextLine = fgetl(fIN);
tableIndex = 0;
tableLine = 0;
csvTable = [];
% start reading the csv file, line by line
while nextLine ~= -1
lineStr = strtrim(strsplit(nextLine,',')) ;
% remove empty cells
lineStr(cellfun('isempty',lineStr)) = [] ;
tableLine = tableLine + 1 ;
% if 1 element start new table
if numel(lineStr) == 1
tableIndex = tableIndex + 1;
tableLine = 1;
csvTable{tableIndex,tableLine} = lineStr ;
else
lineStr = add_comas(lineStr) ;
csvTable{tableIndex,tableLine} = lineStr ;
end
nextLine = fgetl(fIN);
end
fclose(fIN);
fOUT = fopen(filename_OUTPUT,'w');
if row_index > size(csvTable,2) -2
error('The row index exceeds the maximum number of rows!')
end
for k = 1 : size(csvTable,1)
title = csvTable{k,1};
columnHeaders = csvTable{k,2};
selected_row = csvTable{k,row_index+2};
fprintf(fOUT,'%s\n',title{:});
fprintf(fOUT,'%s',columnHeaders{:});
fprintf(fOUT,'\n');
fprintf(fOUT,'%s',selected_row{:});
fprintf(fOUT,'\n');
end
fclose(fOUT);
function line_with_comas = add_comas(this_line)
for ii = 1 : length(this_line)-1
this_line{ii} = strcat(this_line{ii},',') ;
end
line_with_comas = this_line ;
I have the following formula as the grouping for a Cross Tab Report:
{Command.Year} & ' ' & {Command.RF Period}
Year is a SmallInt and Period is a TinyInt.
The problem is that it shows on the report as:
2,009.00 9.00
The database values are actually:
2009 9
I can't remove the decimal places via formatting because they are in the formula together.
Ultimately I'd like it to be:
2009 09
Edit:
I found this link: http://www.kenhamady.com/form15.shtml
Now my code looks like this for period:
WhileReadingRecords;
StringVar text := Totext ( {Command.RF Period} , 6 , "" ) ; //put your numeric field in this line
NumberVar end := length ( text ) ;
NumberVar clip :=
(if Val ( text [ end - 6 to end ] ) = 0 then 1 else 0 ) +
(if Val ( text [ end - 5 to end ] ) = 0 then 1 else 0 ) +
(if Val ( text [ end - 4 to end ] ) = 0 then 1 else 0 ) +
(if Val ( text [ end - 3 to end ] ) = 0 then 1 else 0 ) +
(if Val ( text [ end - 2 to end ] ) = 0 then 1 else 0 ) +
(if Val ( text [ end - 1 to end ] ) = 0 then 1 else 0 ) +
(if Val ( text [ end - 0 to end ] ) = 0 then 1 else 0 ) ;
text [ 1 to Length ( text ) - clip ]
However, I don't use Crystal Language, I use VB. How do I append a 0 in front of the period if it does not begin with a 1?
The problem now is that September (9) shows up after October, Nov, and Dec because aphabetically 9 comes after 1.
Anybody?
The ToText function is very useful for this kind of thing, no loops required. In Crystal's VB Syntax :
Formula = ToText({Command.Year}, 0, "") & " " & ToText({Command.RF Period}, "00")
This should work if {Command.Year} and {Command.RF Period} are integers as you describe.