I'm working with a Ruby script that reads a .json file.
Here is the JSON file:
{
"feed.xml": "93d5b140dd2b4779edef0347ac835fb1",
"index.html": "1cbe25936e392161bad6074d65acdd91",
"md5.json": "655d7c1dbf83a271f348a50a44ba4f6a",
"test.sh": "9be192b1b5a9978cb3623737156445fd",
"index.html": "c064e204040cde216d494776fdcfb68f",
"main.css": "21b13d87db2186d22720e8c881a78580",
"welcome-to-jekyll.html": "01d7c7d66bdeecd9cd69feb5b4b4184d"
}
It is completely valid, and is checked for its existence before trying to read from it. Example:
if File.file?("md5.json")
puts "MD5s exists"
mddigests = File.open("md5.json", "r")
puts "MD5s" + mddigests.read
items = JSON.parse(mddigests.read) <--- Where it all goes wrong.
puts items["feed.xml"]
Everything works up until that point:
MD5s exists
MD5s{
"feed.xml": "93d5b140dd2b4779edef0347ac835fb1",
"index.html": "1cbe25936e392161bad6074d65acdd91",
"md5.json": "655d7c1dbf83a271f348a50a44ba4f6a",
"test.sh": "9be192b1b5a9978cb3623737156445fd",
"index.html": "c064e204040cde216d494776fdcfb68f",
"main.css": "21b13d87db2186d22720e8c881a78580",
"welcome-to-jekyll.html": "01d7c7d66bdeecd9cd69feb5b4b4184d"
}
common.rb:156:in `initialize': A JSON text must at least contain two octets! (JSON::ParserError)
I've searched and tried a lot of different things, to no avail. I'm stumped. Thanks!
You have a duplicate call to read() at the point that it all goes wrong. Replace the second call to read() with the variable mddigests and all should be fine.
This code should work like you'd expect:
if File.file?("md5.json")
puts "MD5s exists"
mddigests = File.open("md5.json", "r")
digests = mddigests.read
puts "MD5s" + digests
items = JSON.parse(digests) #<--- This should work now!
puts items["feed.xml"]
end
The reason is that the file pointer is moved after the first read(), and by the second read(), it's at the end of file, hence the message requiring at least 2 octets.
Related
Silly question, but I want to do some processing on a dataset and put them into different CSVs, like UDID1.csv, UDID2.csv, ..., UDID1000.csv. So this is my code:
for i in 1..1000
logfile = File.new('C:\Users\hp1\Desktop\Datasets\New File\UDID#{i}\.csv',"a")
#I'll do some processing here
end
But the program throws an error when running because of the UDID#{i} part. So, how to overcome this issue? Thanks.
Edit: This is the error:
in `initialize': No such file or directory # rb_sysopen - C:\Users\hp1\Desktop\Datasets\New File\udid#{1}\.csv (Errno::ENOENT)from C:/Ruby21/bin/hashedUDID.rb:38:in `new' from C:/Ruby21/bin/hashedUDID.rb:38:in '<main>'
The ' is one problem, another problem is the path.
In your posting the New File must exist as a directory. Inside this directory must exist another directories like UDID0001. This gets a .csv file.
Correct is (I don't use the non-rubyesk for-loop):
1.upto(1000) do |i|
logfile = File.new("C:\\Users\\hp1\\Desktop\\Datasets\\UDID#{i}.csv", "a")
#I'll do some processing here
logfile.close #Don't forget to close the file
end
Inside " the backslash must be masked (\\). Instead you may use /:
logfile = File.new("C:/Users/hp1/Desktop/Datasets/New File/UDID#{i}/.csv", "a")
Another possibility is the usage of %i to insert the number:
logfile = File.new("C:/Users/hp1/Desktop/Datasets/New File/UDID%02i/.csv" % i, "a")
I prefer to use open, then the file is closed with the end of the block:
File.open("C:/Users/hp1/Desktop/Datasets/New File/UDID%04i/.csv" % i, "a") do |logfile|
#I'll do some processing here
end #closes the file
Warning:
I'm not sure, if you really want to create 1000 log files (The File is opened inside the loop. so each step creates a file.).
If yes, then the %04i-version has the advantage, that the files get all the same number of digits (starting with 0001 and ending with 1000).
(1..10).each { |i| logfile = File.new("/base/path/UDID#{i}.csv") }
You must use double quote (") when you need string interpolation.
#{} can only be used in strings with double quotes ". So change your code to:
for i in 1..1000
logfile = File.new("C:\Users\hp1\Desktop\Datasets\New File\UDID#{i}\.csv","a")
# other stuff
end
I want to open a csv file using SmarterCSV.process
market_csv = SmarterCSV.process(market)
p "just read #{market_csv}"
The problem is that the data is not read and this prints:
[]
However, if I attempt the same thing with the default CSV library implementation the content of the file is read(the following print statement prints the file).
CSV.foreach(market) do |row|
p row
end
The content of the file I was reading is of the form:
Date,Close
03/06/15,0.1634
02/06/15,0.1637
01/06/15,0.1638
31/05/15,0.1638
The problem could come from the line separator, the file is not exactly the same if you're using windows or unix system ("\r\n" or "\r"). Try to identify and specify the character in the SmarterCSV.process like this:
market_csv = SmarterCSV.process(market, row_sep: "\r")
p "just read #{market_csv}"
or like this:
market_csv = SmarterCSV.process(market, row_sep: :auto)
p "just read #{market_csv}"
I have a file that is listed line by line as such:
first.last#example.com
first.last#example.com
last#example.com...
Note that some of the addresses don't have a first name, in which case, it is just the last name.
How can I write a simple Ruby script to read in this file (call it email.txt)
and write it back to the file in sorted order by last name?
Put this in a file, e.g. sort_by_last.rb:
puts IO.readlines('email.txt').sort_by { |e| e.match(/[^\.]+(?=#)/)[0].downcase }
then run it:
ruby sort_by_last.rb > emails_sorted.txt
For variable filename
Set contents of sort_by_last.rb to
puts STDIN.readlines.sort_by { |e| e.match(/[^\.]+(?=#)/)[0].downcase }
then run:
ruby sort_by_last.rb < email.txt > emails_sorted.txt
I am trying to read the contents of the file from a local disk as follows :
content = File.read("C:\abc.rb","r")
when I execute the rb file I get an exception as Error: No such file or directory .What am I missing in this?
In a double quoted string, "\a" is a non-printable bel character. Similar to how "\n" is a newline. (I think these originate from C)
You don't have a file with name "C:<BEL>bc.rb" which is why you get the error.
To fix, use single quotes, where these interpolations don't happen:
content = File.read('C:\abc.rb')
content = File.read("C:\/abc.rb","r")
First of all:
Try using:
Dir.glob(".")
To see what's in the directory (and therefore what directory it's looking at).
open("C:/abc.rb", "rb") { |io| a = a + io.read }
EDIT: Unless you're concatenating files together, you could write it as:
data = File.open("C:/abc.rb", "rb") { |io| io.read }
I have a requirement where I need to replace the element value with the new one and I dont want any other modification to be done to the file.
<mtn:test-case title='Power-Consist-Message'>
<mtn:messages>
<mtn:message sequence='4' correlation-key='0x0F04'>
<mtn:header>
<mtn:protocol-version>0x4</mtn:protocol-version>
<mtn:message-type>0x0F04</mtn:message-type>
<mtn:message-version>0x01</mtn:message-version>
<mtn:gmt-time-switch>false</mtn:gmt-time-switch>
<mtn:crc-calc-switch>1</mtn:crc-calc-switch>
<mtn:encrypt-switch>false</mtn:encrypt-switch>
<mtn:compress-switch>false</mtn:compress-switch>
<mtn:ttl>999</mtn:ttl>
<mtn:qos-class-of-service>0</mtn:qos-class-of-service>
<mtn:qos-priority>2</mtn:qos-priority>
<mtn:qos-network-preference>1</mtn:qos-network-preference>
this is how the xml file looks like, I want to replace 999 with "some other value", under s section, but when am doing that using formatter in ruby some other unwanted modifications are taking place, the code that am using is as belows
File.open(ENV['CadPath1']+ "conf\\cad-mtn-config.xml") do |config_file|
# Open the document and edit the file
config = Document.new(config_file)
testField=config.root.elements[4].elements[11].elements[1].elements[1].elements[1].elements[11]
if testField.to_s.match(/<mtn:qos-network-preference>/)
test=config.root.elements[4].elements[11].elements[1].elements[1].elements[1].elements[8].text="2"
# Write the result to a new file.
formatter = REXML::Formatters::Default.new
File.open(ENV['CadPath1']+ "conf\\cad-mtn-config.xml", 'w') do |result|
formatter.write(config, result)
end
end
end
when am writting the modifications to the new file, the xml file size is getting changed from 79kb to 78kb, is there any way to just replace the particular line in xml file and save changes without affecting the xml file.
Please let me know soon...
I prefer Nokogiri as my XML/HTML parser of choice:
require 'nokogiri'
xml =<<EOT
<mtn:test-case title='Power-Consist-Message'>
<mtn:messages>
<mtn:message sequence='4' correlation-key='0x0F04'>
<mtn:header>
<mtn:protocol-version>0x4</mtn:protocol-version>
<mtn:message-type>0x0F04</mtn:message-type>
<mtn:message-version>0x01</mtn:message-version>
<mtn:gmt-time-switch>false</mtn:gmt-time-switch>
<mtn:crc-calc-switch>1</mtn:crc-calc-switch>
<mtn:encrypt-switch>false</mtn:encrypt-switch>
<mtn:compress-switch>false</mtn:compress-switch>
<mtn:ttl>999</mtn:ttl>
<mtn:qos-class-of-service>0</mtn:qos-class-of-service>
<mtn:qos-priority>2</mtn:qos-priority>
<mtn:qos-network-preference>1</mtn:qos-network-preference>
EOT
Notice that the XML is malformed, i.e., it doesn't terminate correctly.
doc = Nokogiri::XML(xml)
I'm using CSS accessors to find the ttl node. Because of some magic, Nokogiri's CSS ignores XML name spaces, simplifying finding nodes.
doc.at('ttl').content = '1000'
puts doc.to_xml
# >> <?xml version="1.0"?>
# >> <test-case title="Power-Consist-Message">
# >> <messages>
# >> <message sequence="4" correlation-key="0x0F04">
# >> <header>
# >> <protocol-version>0x4</protocol-version>
# >> <message-type>0x0F04</message-type>
# >> <message-version>0x01</message-version>
# >> <gmt-time-switch>false</gmt-time-switch>
# >> <crc-calc-switch>1</crc-calc-switch>
# >> <encrypt-switch>false</encrypt-switch>
# >> <compress-switch>false</compress-switch>
# >> <ttl>1000</ttl>
# >> <qos-class-of-service>0</qos-class-of-service>
# >> <qos-priority>2</qos-priority>
# >> <qos-network-preference>1</qos-network-preference>
# >> </header></message></messages></test-case>
Notice that Nokogiri replaced the content of the ttl node. It also stripped the XML namespace info because the document didn't declare it correctly, and, finally, Nokogiri has added closing tags to make the document syntactically correct.
If you want the namespace to be declared in the output, you'll need to make sure it's there in the input.
If you need to just literally replace that value without affecting anything else about the XML file, even if (as pointed by the Tin Man above) that would mean leaving the original XML file malformed, you can do that with direct string manipulation using a regular expression.
Assuming there is guaranteed to only be one <mtn:ttl> tag in your XML document, you could just do:
doc = IO.read("somefile.xml")
doc.sub! /<mtn:ttl>.+?<\/mtn:ttl>/, "<mtn:ttl>some other value<\/mtn:ttl>"
File.open("somefile.xml", "w") {|fh| fh.write(doc)}
If there might be more than one <mtn:ttl> tag, then this is trickier; how much trickier depends on how you want to figure out which tag(s) to change.