CSV processing and mutual conversion of web related data types

base64, csirt, csv, e-mail, spreadsheet, whois
pip install convey==1.3.9rc3



Build Status

Swiss knife for mutual conversion of the web related data types, like base64 or outputs of the programs whois, dig, curl. Convenable way to quickly gather all meaningful information or to process large files that might freeze your spreadsheet processor.

Any input is accepted:

  • if a single value input is detected, all meaningful information is fetched
  • multiline base64/quoted_printable string gets decoded
  • log file is converted to CSV
  • CSV file (any delimiter, header or whatever) performs one or more actions
    1. Pick, delete or sort columns (if only some columns are needed)
    2. Add a column (computes one field from another – see below)
    3. Filter (keep/discard rows with specific values, no duplicates)
    4. Split by a column (produce separate files instead of single file; these can then be sent by generic SMTP or through OTRS)
    5. Change CSV dialect (change delimiter or quoting character, remove header)
    6. Aggregate (count grouped by a column, sum...)

Python3.6+ required.

Table of contents


Usage 1 – Single query

Check what happens if an IP is provided, it returns table with WHOIS-related information and scraped HTTP content.

$ convey # single query input
Input value detected: ip ...au

field             value
----------------  -----------------
asn               as13335
abusemail         abuse@apnic.net
country           au
netname           apnic-labs
csirt-contact     -
incident-contact  au
status            200
text              DNSThe free app that makes your (much longer text...)

Usage 2 – CSV processor program

Parses CSV file.

$ convey my-file-with-ips.csv # will trigger file parsing
Source file: /tmp/my-file.csv
Log lines: 200


Delimiter character found: ','
Quoting character: '"'
Header is present: not used

Could you confirm this? [y]/n

Usage 3 – Web service

Again, let's provide an IP to the web service, it returns JSON with WHOIS-related information and scraped HTTP content.

$ convey --server  # start a UWSGI session

Access in the browser: http://localhost:26683/?q=example.com

{"ip": "", "prefix": "", "asn": "", "abusemail": "abuse@verizondigitalmedia.com", "country": "unknown", "netname": "edgecast-netblk-03", "csirt-contact": "-", "incident-contact": "unknown", "status": 200, "text": "DNSThe free app that makes your (much longer text...)"}

Installation and first run


  • You'll be asked to install dialog library at the first run if not already present in the system.
  • If something is missing on your system, you may find help yourself with this command:
    sudo apt install python3-pip python3-dev python3-tk git xdg-utils dialog whois dnsutils nmap curl build-essential libssl-dev libpcre3 libpcre3-dev swig && pip3 install setuptools wheel uwsgi && pip3 install --upgrade ipython
    • build-essential is needed to build uwsgi and envelope
    • libpcre3 libpcre3-dev needed to suppress uWSGI warning !!! no internal routing support, rebuild with pcre support !!!
    • libssl-dev needed to be present before building uwsgi if you will need to use --https
    • python3-dev swig needed to build envelope

Launch as a package:

# (optional) setup virtual environment
python3 -m venv venv
. venv/bin/activate
(venv) $ ... # continue below

# install from PyPi
pip3 install convey  # without root use may want to use --user

# (optional) alternatively, you may want to install current master from GitHub
pip3 install git+https://github.com/CZ-NIC/convey.git

# launch
convey [filename or input text] # or try `python3 -m convey` if you're not having `.local/bin` in your executable path

Parameter [filename or input text] may be the path of the CSV source file or any text that should be parsed. Note that if the text consist of a single value, program prints out all the computable information and exits; I.E. inputting a base64 string will decode it.

OR launch from a directory

# download from GitHub
git clone git@github.com:CZ-NIC/convey.git
cd convey
pip3 install -r requirements.txt  --user

# launch

Bash completion

  1. Run: apt-get install bash-completion jq
  2. Copy: extra/convey-autocompletion.bash to /etc/bash_completion.d/
  3. Restart terminal


  • Convey tries to open the file in the default GUI editor or in the terminal editor if GUI is not an option.
  • If config.ini is present at working directory, that one is used over the one in the user config folder.
  • Configuration is updated automatically on upgrade.

Computing fields

Computable fields

Some of the field types we are able to compute:

  • abusemail – got abuse e-mail contact from whois
  • asn – got from whois
  • base64 – encode/decode
  • country – country code from whois
  • csirt-contact – e-mail address corresponding with country code, taken from your personal contacts_abroad CSV in the format country,abusemail. Path to this file has to be specified in config.ini » contacts_abroad
  • external – you specify method in a custom .py file that receives the field and generates the value for you, see below
  • hostname – domain from url
  • incident-contact – if the IP comes from local country (specified in config.ini » local_country) the field gets abusemail, otherwise we get country. When splitting by this field, convey is subsequently able to send the split files to local abuse and foreign csirt contacts
  • ip – translated from url
  • netname – got from whois
  • prefix – got from whois

Detectable fields

Some of the field types we are able to auto-detect:

  • ip – standard IPv4 / IPv6 addresses
  • cidr – CIDR notation, ex:
  • port_ip – IPv4 in the form
  • any_ip – IPv4 garbled in the form any text any text
  • hostname – or FQDN; 2nd or 3rd domain name
  • url – URL starting with http/https
  • asn – AS Number
  • base64 – text encoded with base64
  • wrong_url – URL that has been deactivated by replacing certain chars, ex: "hxxp://example[.]com"

Overview of all methods:

Current field computing capacity can be get from --show-uml flag. Generate yours by ex: convey --show-uml | dot -Tsvg -o /tmp/convey-methods.svg

  • Dashed node: field type is auto-detectable
  • Dashed edge: field types are identical
  • Edge label: generating options asked at runtime
  • Rectangle: field category border

Methods overview

External field how-to

Simple custom method

If you wish to compute an external field, prepare a file whose contents can be as simple as this:

def any_method(value):
    # do something
    return "modified :)"
Launch an external method
  • When CSV processing, hit 'Add column' and choose 'new external... from a method in your. py file'
  • Or in the terminal append --field external to your convey command. A dialog for a path of the Python file and desired method will appear.
$ convey [string_or_filepath] --field external 
  • You may as well directly specify the path and the callable. Since the --field has following syntax:


You may omit both COLUMN and SOURCE_TYPE writing it this way:


$ convey [string_or_filepath] --field external,/tmp/myfile.py,any_method
Input value seems to be plaintext.
field     value
--------  -----------------------
external  modified :)
Register an external method
  • You may as well hard code custom fields in the config.ini by providing paths to the entry point Python files delimited by a comma: external_fields = /tmp/myfile.py, /tmp/anotherfile.py. All the public methods in the defined files will become custom fields!
external_fields = /tmp/myfile.py
  • If this is not needed, you may register one by one by adding new items to the EXTERNAL section. Delimit the method name by a colon.
any_method = /tmp/myfile.py:any_method

List of results possible

If you need a single call to generate multiple rows, return list, the row accepting a list will be duplicated.

def any_method(value):
    # do something
    return ["foo", "bar"]

When convey receives multiple lists, it generates a row for each combination. Ex: If a method returns 2 items and another 3 items, you will receive 6 similar rows.

PickMethod decorator

Should there be multiple ways of using your generator, place them as methods of a class decorated with PickMethod and let the user decide at the runtime. PickMethod has optional default:str parameter that specifies the default method.

from convey import PickMethod

class any_method(PickMethod):
    def all(x):
        ''' All of them.  '''
        return x
    def filtered(cls, x):
        ''' Filter some of them '''
        if x in country_code_set:
            return x
$ convey file.csv --field any_method  # user will be asked whether to use `all` or `filtered`
$ convey file.csv --field any_method[filtered]  # `filtered` sub-method will be used
$ convey file.csv --field any_method --yes  # the default `all` sub-method will be used

PickInput decorator

If you need a direct user entry before each processing, import PickInput and make your method accept two parameters. The latter will be set by the user and may have a default value.

from convey import PickInput

def time_format(val, format="%H:%M"):
    ''' This text will be displayed to the user.
        If running in headless mode, the default format will be "%H:%M" (hours:minutes).   '''
    return dateutil.parser.parse(val).strftime(format)
$ convey file.csv --field time_format  # user will be asked for a format parameter
$ convey file.csv --field time_format[%M]  # `format` will have the value `M%`
$ convey file.csv --field time_time --yes  # the default `format` `%H:%M` will be used

Web service

When launched as a web service, three parameters are available:

  • q – search query

  • type – same as --type CLI flag

  • field – same as --field CLI flag.

    • Note that unsafe field types code and external are disabled in web service by default. You may re-allow them in webservice_allow_unsafe_fields config option
    • full syntax of CLI flag is supported


    Ex: reg_s,l,L performs regular substitution of 'l' by 'L'

Quick deployment may be realized by a single command:

$ convey --server

Internally, flag --server launches wsgi.py with a UWSGI session.

  • Note that convey must be installed via pip
  • Note that LACNIC may freeze for 300 s, hence the timeout recommendation.
  • Note that you may find your convey installation path by launching pip3 show convey
$ uwsgi --http :26683 --http-timeout 310 --wsgi-file /home/$USER/.local/lib/python3.7/site-packages/convey/wsgi.py

Access: curl http://localhost:26683/?q=example.com

{"ip": "", "prefix": "", "asn": "", "abusemail": "abuse@verizondigitalmedia.com", "country": "unknown", "netname": "edgecast-netblk-03", "csirt-contact": "-", "incident-contact": "unknown", "status": 200, "text": "DNSThe free app that makes your (much longer text...)"}

Access: curl http://localhost:26683/?q=example.com&field=ip

{"ip": ""}

Access: curl http://localhost:26683?q=hello&type=country&field=reg_s,l,L

{"reg_s": "heLLo"}

Sending files

When you split the CSV file into chunks by an e-mail, generated files may be sent to these addresses. Look at the example of an unlocked "send" menu below. You see the list of the recipients, followed by a conditional list of recipients that have been already sent to. Next, an exact e-mail message is printed out, including headers.

In the menu, you may either:

  • Send the e-mails
  • Limit the messages that are being send at once; if you are not 100 % sure you want to send the the whole message bucket at once.
  • Edit the template. The message file will open either in the default GUI or terminal editor. The first line of the template should be Subject: ..., followed by a free line. Note that you may include any e-mail headers, such as Reply-To: ..., Cc: ..., etc. The e-mail will reflect all of them. You may write the message either in plain text or in the HTML.
  • Choose which recipients in a checkbox list will receive the message.
  • Test sending a message to your own address. You'll be prompted which of the messages should be delivered to you. The e-mail contents possibly modified by a dynamic template is shown just before sending.
  • Print all e-mails to a file to have the more granulated control over what is going to be sent.
  • Toggle file attaching on and off. There are use cases when you do not want the files to be sent with, the text suits fine.
  *** E-mail template ***
Recipient list (1/3): alice@example.com
Already sent (2/3): bob@example.com, cilia@example.com

Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
MIME-Version: 1.0
Subject: My subject
From: me@example.com
Date: Fri, 17 Jan 2020 01:36:28 +0100


this is my testing message.

Keen regards

1) Send all e-mails (1) ←←←←←
l) Limit sending amount to...
e) Edit template...
r) Choose recipients...
t) Send test e-mail...
p) Print e-mails to a file...
a) Attach files (toggle): True
x) Go back...

Arbitrary e-mail headers, "From" header, GPG signing

In the template, you may specify any e-mail header, such as Reply-To, Cc or From. If From is not found, we take SMTP/email_from_name config value. If gnupg home is found on the default user path, we check if there is a secret key matching the From header and if found, e-mail will be GPG-signed. If it is going to be signed, you would see something like Content-Type: multipart/signed; protocol="application/pgp-signature"; header in the e-mail template preview.

Dynamic templates

Message is processed with Jinja2 templating system by default.

Few instruments are included to treat the attachment contents:

  • row() – Generate attachment contents fields row by row. Header skipped.
  • print_attachment() – Prints the attachment contents and prevent it to be attached.
    You will find our findings below.
    {{ print_attachment() }}
  • amount(count=2) – Check if the attachment has at least count number of lines. Header is not counted. Useful when deciding whether the are single row in the result or multiple.
  • joined(column: int, delimiter=", ") – Return a column joined by delimiter
  • first_row – Access first line fields Example:
    {% if amount() %}
        Here is the complete list of the elements.    
        {{ joined(1,"\n") }}
    {% else %}
        Here is the element you had problems with: {{ first_row[1] }}
    {% endif %}


In the examples, we will use these parameters to add a field and to shorten the result.

# -f, --field adding field syntax: FIELD[[CUSTOM]],[COLUMN],[SOURCE_TYPE],[CUSTOM],[CUSTOM]
# -H, --headless: just quietly print out single value, no dialog

URL parsing

Output formats

Put any IP or URL as the argument.

$ convey example.com
Input value detected: hostname

Whois abuse@verizondigitalmedia.com
Scrapping http://example.com...
field             value
----------------  ------------------------------------------------------------------------------
tld               com
url               http://example.com
abusemail         abuse@verizondigitalmedia.com
csirt_contact     -
incident_contact  abuse@verizondigitalmedia.com
netname           edgecast-netblk-03
aaaa              2606:2800:220:1:248:1893:25c8:1946
mx                0 .
ns                ['a.iana-servers.net.', 'b.iana-servers.net.']
spf               v=spf1 -all
http_status       200
text              Example Domain
                  This domain is for use in illustrative examples in documents. You may use this
                   domain in literature without prior coordination or asking for permission.
                  More informatio

Should you need just the country the domain/IP is hosted in, use --field, -f argument

$ convey wikipedia.com -f country
Input value detected: hostname

Whois us
field    value
-------  -------
country  us

Use --headless, -H or --quiet, -q flag to shorten the output (and cut down all dialogues).

$ convey wikipedia.com -f country -H

Flag --json modifies the output.

$ convey wikipedia.com -f country -H --json
{"country": "us"}

Computing TLD from another column

To compute a TLD from the abusemail that is being used for the IP domain is hosted in, add a field abusemail and then another field tld. Specifically say that the latter should source from the second column (which is abusemail) – either type '2' or 'abusemail'.

$ convey example.com -f abusemail -f tld,2
$ convey example.com -f abusemail -f tld,abusemail
Input value detected: hostname

Whois abuse@verizondigitalmedia.com
field      value
---------  -----------------------------
abusemail  abuse@verizondigitalmedia.com
tld        com

To prevent abusemail from being output, use --field-excluded, -fe instead of --field, -f:

$ convey example.com -fe abusemail -f tld,2 -H
Input value detected: hostname

Whois abuse@verizondigitalmedia.com
field    value
-------  -------
tld      com

We did not say earlier, user is asked each time whether they wish to get any tld, gTLD (ex: com) or ccTLD (ex: cz). You may specify it from CLI by one of those equivalent commands.

$ convey test.csv --fresh --field tld[gTLD]
$ convey test.csv --fresh --field tld,,,gTLD

# flag --yes or --headless will choose the default option which is *all*
$ convey test.csv --fresh --field tld --yes

CSV processing

Should you have a list of the object that you want to enrich of a CIDR they are hosted at, load the file test.csv they are located in.

# file text.csv
domain list

And see the menu just by adding --field cidr argument.

$ convey test.csv -f cidr
Source file: /tmp/ram/test.csv
Identified columns: 
Log lines: 3

domain list

Delimiter character found: ','
Quoting character: '"'
Header is present: yes

Could you confirm this? [y]/n: (HIT ENTER)

Source file: /tmp/ram/test.csv, delimiter: ',', quoting: '"', header: used
Identified columns: domain list (hostname)
Computed columns: cidr (from domain list)
Log lines: 3

domain list

Whois us
Whois abuse@verizondigitalmedia.com
domain list      cidr from:
   (hostname)    domain list
---------------  ---------------

Main menu - how the file should be processed?
1) Pick or delete columns
2) Add a column
3) Filter
4) Split by a column
5) Change CSV dialect
6) Aggregate
p) process ←←←←←
~) send (split first)
~) show all details (process first)
r) reset...
c) config...
x) exit

File splitting

We will create an ASN field and split the file.csv by this field, without adding it into the output.

# file.csv
$ convey file.csv --field-excluded asn --split asn
** Processing completed: 3 result files in /tmp/ram/file.csv_convey1573236314
# file as14907
# file as43037
# file as15169

CSIRT Usecase

A CSIRT may use the tool to automate incident handling tasks. The input is any CSV we receive from partners; there is at least one column with IP addresses or URLs. We fetch whois information and produce a set of CSV grouped by country AND/OR abusemail related to IPs. These CSVs are then sent by through OTRS from within the tool.
A most of the work is done by this command.

convey --field-excluded incident_contact,source_ip --split incident_contact --yes [FILENAME]

Custom code field

Adding a column from custom Python code:

$ convey example.com -f code,"x=x[1:5]"

Base64 and Regular expressions

Code there and back:

$ convey hello -f base64  -H  # --headless conversion to base64
$ convey aGVsbG8= -H  # automatically identifies input as base64 and produces plaintext

Use a reg column for regular expressions.

# start adding a new reg column wizzard that will take decoded "hello" as input 
$ convey aGVsbG8= -f reg
$ convey aGVsbG8= -f reg_s,"ll","LL" -H   # substitute 'll' with 'LL'

Specify source

# start adding a new reg column wizzard that will take plaintext "aGVsbG8=" as input 
$ convey aGVsbG8= -f reg,plaintext
# specifying plaintext as a source type will prevent implicit convertion from base64
$ convey aGVsbG8= -f reg_s,plaintext,"[A-Z]","!" -H  # substitute uppercase letters with '!'

Converting units

We are connected to the pint unit converter!

$ convey "3 kg" 
Input value detected: unit

field      value
---------  --------------------------------------------------------------------------------------------
plaintext  ['1.806642538265029e+27 atomic_mass_unit', '105.82188584874123 ounce', '96.45223970588393 ap
           othecary_ounce', '0.0703602964419822 bag', '0.05905239165666364 long_hunderweight', '0.06613
           867865546327 US_hundredweight', '0.002952619582833182 UK_ton', '0.002952619582833182 long_to
           n', '1929.0447941176785 pennyweight', '46297.07505882429 grain', '1.7935913792661326e+27 pro
           ton_mass', '771.6179176470714 apothecary_dram', '3000.0 gram', ...]

$ convey "3 kg" -f unit # launches wizzard that let's you decide what unit to convert to 
$ convey "3 kg" -f unit[g] -H
3000.0 gram

$ convey "kg" -f unit --csv-processing --headless
kg|6.022141794216764e+26 atomic_mass_unit
kg|0.001 metric_ton
kg|0.0009842065276110606 UK_ton
kg|771.6179176470714 scruple
kg|257.2059725490238 apothecary_dram
kg|1000.0 gram

# You may try to specify the units with no space and quotation.
# In the following example, convey expand all time-units it is able to compute
# – time units will be printed out and each is base64 encoded. 
$ convey 3hours
Input value detected: timestamp, unit

field                value
-------------------  ------------------------------------------------------------------
base64               ['MC4wMDQxMDY4NjM4OTc0NTAwNzcgbW9udGg=', 'MTA4MDAuMCBzZWNvbmQ=', 'MC4wMTc4NTcxNDI4NTcxNDI4NSB3ZWVr'] 
plaintext            ['0.004106863897450077 month', '10800.0 second', '0.01785714285714285 week', (...)]
time                 03:00:00

# What if
$ convey 3hours -f urlencode
Input value detected: timestamp, unit

Input unit_expand variable unit: *you type here sec or seconds to see the wizzard*                                                                                                                                                                                                                                           
| original   | result         |                                                                                                                                                                                    
| 3hours     | 10800.0 second |                                                                                                                                                                                    

field      value
---------  ----------------
urlencode  10800.0%20second

# What if we wanted to urlencode text "3hours" without converting it to unit first? 
# Just specify the SOURCE_TYPE to be plaintext:
$ convey "3hours" -f urlencode,plaintext
Input value detected: timestamp, unit

field      value
---------  -------
urlencode  3hours

Country names

When specifying country code, you get country name.

$ convey --type country sc

Lots of countries can be recognized.

$ convey --type country_name Futuna

You may get the country code from various telephone number formats.

$ convey +2481234567

$ convey "1-541-754-3010"
['ca', 'us']

$ convey 


Syntax is [COLUMN, FUNCTION], ..., [group-by-COLUMN].
Possible functions are:

  • avg
  • sum
  • count
  • min
  • max
  • list
  • set

Let's have a file.

# file.csv

Sum the price column.

$ convey file.csv --aggregate price,sum

Group the price sum by category.

$ convey file.csv --aggregate price,sum,category
category     sum(price)
---------  ------------
total               972
bulb                370
kettle              602

Group the price sum and the consumption average value by category.

$ convey file.csv --aggregate price,sum,consumption,avg,category
category      sum(price)    avg(consumption)
----------  ------------  ------------------
total                972               41
bulb                 370               18.33
kettle               602               75

Group the price sum by category and list its values.

$ convey file.csv --aggregate price,sum,price,list,category
category      sum(price)  list(price)
----------  ------------  ---------------------
total                972  (all)
bulb                 370  ['100', '150', '120']
kettle               602  ['250', '352']

You can even split while aggregating. Each file will count its own results.

$ convey file.csv --agg price,sum --split category

Split location: bulb

Split location: kettle


Brought by CSIRT.cz.