Is it possible how to reverse Zipcode using shellscript? i am try to get zipcode from latitude and longitude so should is it possible?
location=$(curl -s http://freegeoip.net/csv/$1)
echo $location
RTFM, http:freegeoip.net/ says
freegeoip.net/{format}/{IP_or_hostname}
This does not support giving it a lat,lon location.
google: reverse geocode postal code
Has lots of solutions.
Related
I have a bunch of instances running in GCE. I want to programmatically get a list of the internal IP addresses of them without logging into the instances (locally).
I know I can run:
gcloud compute instances list
But are there any flags I can pass to just get the information I want?
e.g.
gcloud compute instances list --internal-ips
or similar? Or am I going to have to dust off my sed/awk brain and parse the output?
I also know that I can get the output in JSON using --format=json, but I'm trying to do this in a bash script.
The simplest way to programmatically get a list of internal IPs (or external IPs) without a dependency on any tools other than gcloud is:
$ gcloud --format="value(networkInterfaces[0].networkIP)" compute instances list
$ gcloud --format="value(networkInterfaces[0].accessConfigs[0].natIP)" compute instances list
This uses --format=value which also requires a projection which is a list of resource keys that select resource data values. For any command you can use --format=flattened to get the list of resource key/value pairs:
$ gcloud --format=flattened compute instances list
A few things here.
First gcloud's default output format for listing is not guaranteed to be stable, and new columns may be added in the future. Don't script against this!
The three output modes are three output modes that are accessible with the format flag, --format=json, --format=yaml, and format=text, are based on key=value pairs and can scripted against even if new fields are introduced in the future.
Two good ways to do what you want are to use JSON and the jq tool,
gcloud compute instances list --format=json \
| jq '.[].networkInterfaces[].networkIP'
or text format and grep + line-oriented using tools,
gcloud compute instances list --format=text \
| grep '^networkInterfaces\[[0-9]\+\]\.networkIP:' | sed 's/^.* //g'
I hunted around and couldn't find a straight answer, probably because efficient tools weren't available when others replied to the original question. GCP constantly updates their libraries & APIs and we can use the filter and projections to extract targeted attributes.
Here I outline how to reserve an external static IP, see how it's attributes are named & organised, and then export the external IP address so that I can use it in other scripts (e.g. assign this to a VM instance or authorise this network (IP address) on a Cloud SQL instance.
Reserve a static IP in a region of your choice
gcloud compute --project=[PROJECT] addresses create [NAME] --region=[REGION]
[Informational] View the details of the regional static IP that was reserved
gcloud compute addresses describe [NAME] --region [REGION] --format=flattened
[Informational] List the attributes of the static IP in the form of key-value pairs
gcloud compute addresses describe [NAME] --region [REGION] --format='value(address)'
Extract the desired value (e.g. external IP address) as a parameter
export STATIC_IP=$(gcloud compute addresses describe [NAME] --region [REGION] --format='value(address)’)
Use the exported parameter in other scripts
echo $STATIC_IP
The best possible way would be to have readymade gcloud command use the same as and when needed.
This can be achieved using table() format option with gcloud as per below:
gcloud compute instances list --format='table(id,name,status,zone,networkInterfaces[0].networkIP :label=Internal_IP,networkInterfaces[0].accessConfigs[0].natIP :label=External_IP)'
What does it do for you?
Get you data in clean format
Give you option to add or remove columns
Need additional columns? How to find column name even before you run the above command?
Execute the following, which will give you data in raw JSON format consisting value and its name, copy those names and add them into your table() list. :-)
gcloud compute instances list --format=json
Plus Point: This is pretty much same syntax you can tweak with any GCP resources data to fetch including with gcloud, kubectl etc.
As far as I know you can't filter on specific fields in the gcloud tool.
Something like this will work for a Bash script, but it still feels a bit brittle:
gcloud compute instances list --format=yaml | grep " networkIP:" | cut -c 14-100
I agree with #Christiaan. Currently there is no automated way to get the internal IPs using the gcloud command.
You can use the following command to print the internal IPs (4th column):
gcloud compute instances list | tail -n+2 | awk '{print $4}'
or the following one if you want to have the pair <instance_name> <internal_ip> (1st and 4th column)
gcloud compute instances list | tail -n+2 | awk '{print $1, $4}'
I hope it helps.
First time poster so my apologies if this has been covered in a previous topic that I was unable to locate. Basically I'm tasked with creating a script to perform a NSLookup on 50 domain names, format the results and pass them to the hosts file. I'll worry about checking and overwriting duplicate entries later.
Example:
Input: nslookup www.cbc.ca
Result:
Name: a1849.gc.akamai.net
Addresses: 184.50.238.64, 184.50.238.89
Aliases: www.cbc.ca, www.cbc.ca.edgesuite.net
Eventual Output: #184.50.238.64 www.cbc.ca a1849.gc.akamai.net
I figured this was possible with grep, awk and sed but have been messing about with switches and haven't gotten the right combination (mostly cause I'm not the most learned when it comes to regular expressions.) I'm partial to vbs, batch, cmd suggestions as well.
Thanks in advance for the time and effort! :)
nslookup $NAME | awk -v name="$NAME" 'BEGIN{hit=0; addr=""; alias=""} /answer:/{hit=1} /^Address:/{if (hit == 1 && "" == addr) addr=$2} /^Name:/{alias=alias " " $2} END{print(addr, name, alias)}'
Only one address and would not solve multiple identical names like nslookup google.com...
I had a look at the BigQuery command line tool documentation and I saw that you are able to use timestamp literals in a WHERE clause. The documentation shows the following example:
$ bq query "SELECT name, birthday FROM dataset.table WHERE birthday <= '1959-01-01 01:02:05'"
Waiting on job_6262ac3ea9f34a2e9382840ee11538ef ... (0s) Current status: DONE
+------+---------------------+
| name | birthday |
+------+---------------------+
| kim | 1958-06-24 12:18:35 |
+------+---------------------+
As the dataset.table is not a public dataset, I build an example using the wikipedia dataset.
SELECT title, timestamp, SEC_TO_TIMESTAMP(timestamp) AS human_timestamp
FROM publicdata:samples.wikipedia
HAVING human_timestamp>'2008-01-01 01:02:03' LIMIT 5
The example works on the BigQuery Browser but it does not on the bq tool. Why? I tried to use scape characters and several combinations of single and double quotes without success. It is a Windows issue? Here goes a screenshot:
EDIT: This is BigQuery CLI 2.0.18
I know that "It works on my machine" isn't a satisfying answer, but I've tried this on my Mac and on a windows machine, and it appears to work fine on both. Here is the output from my windows machine for the same query you've specified:
C:\Users\Jordan Tigani>bq query "SELECT title, timestamp, SEC_TO_TIMESTAMP(timestamp) AS human_timestamp FROM publicdata:samples.wikipedia HAVING human_timestamp>'2008-01-01 01:02:03' LIMIT 5"
Waiting on bqjob_r607b7a74_00000144b71ddb9b_1 ... (0s) Current status: DONE
Can you make sure that the quotes you're using aren't pasted smart quotes and there aren't any stray unicode characters that might confuse the parsing?
One other hint is to use the --apilog=- option, which tells BigQuery to print out all interaction with the server to stdout. You can then see exactly what is getting sent to the BigQuery backend, and verify that the quotes are as expected.
I found out that the problem is due to the greater operator > in the Windows command line. It does not have anything to do with the google-cloud-sdk, sorry.
It seems that you have to use the scape to echo the sign in the command line: ^>
I found it at google groups (by Todd and Margo Chester), and the official reference at Microsoft site.
So I just started learning UNIX yesterday, and I'm trying to create a basic script that asks for your contact details (name, address, phone number), and then stores that into a file called details.out.
This is driving me NUTS! Its such an easy/basic thing, yet I cant do it, and I've been stuck on it for a solid hour now...
after much googling and searching, I still can't find the answer. So this is what I've done so far, and was wondering where I am going wrong!
echo Please type your first and last name
read $firstname $lastname
echo Please type in your address
read $address
echo Please type in your phone number
read $phone
echo Thank you very much!
echo The details have been stored in '"details.out"'
cat >> details.out <<EOF
Name: echo $firstname echo $lastname
Address: echo $address
Phone Number: echo $phone
EOF
When I read "details.out" it it displays as follows:
Name: echo
Address: echo
Phone Number: echo
ANY help would be appreciated! (and if you get try and point me in the right directions as opposed to straight up giving me the answer, I would appreciate that!)
P.S I'm using Putty if that helps!
when you use read (or declaring variables), don't put $ sigil on the variable names
when you display a variable, always put double quotes around : ex. echo "$var"
when you use here-doc, no need to put echo command
when you use echo, use quotes :
"Double quote" every expansion, and anything that could contain a special character, eg. "$var", "$#", "${array[#]}", "$(command)". Use 'single quotes' to make something literal, eg. 'Costs $5 USD'. See http://mywiki.wooledge.org/Quotes http://mywiki.wooledge.org/Arguments and http://wiki.bash-hackers.org/syntax/words
Whenever you put a $ before a variable name, you're retrieving the current value of that variable. You don't want to do that in your read command. The variables are empty when the script starts, the empty values are put in place of the $firstname and $lastname and read is called with no arguments, causing it to read a line and discard it.
Setting a variable with assignment:
var=value
Setinng a variable with read:
read var
Neither of them use $var because they don't want to look at the current value, they want to replace it.
There's no need for those echos in the heredoc either. They aren't in command position, so they'll just get copied as part of the input to cat.
I need to loop a oracle sqlplus query using bash.
my scenario is like this. I have a set of names in a text file and i need to find out details of that names using a sqlplus query.
textfile.txt content:
john
robert
samuel
chris
bash script
#!/bin/bash
while read line
do
/opt/oracle/bin/sqlplus -s user#db/password #query.sql $line
done < /tmp/textfile.txt
sql query: query.sql
set verify off
set heading off
select customerid from customers where customername like '%&1%';
exit
problem is when I run the script I get errors like
SP2-0734: unknown command beginning
"robert..." - rest of line ignored.
can someone tell me how to solve this?
The way I do this all the time is as follows:
#!/bin/bash
cat textfile.txt |while read Name
do
sqlplus -s userid/password#db_name > output.log <<EOF
set verify off
set heading off
select customerid from customers where customername like '%${Name}%'
/
exit
EOF
Bash will auto magically expand ${Name} for each line and place it into the sql command before sending it into sqlplus
Do you have set define on ? Is your wildcard & ? You could check glogin.sql to know.
And yes, establishing n connections to pass n queries is probably not a good solution. Maybe it's faster for you to develop and you will do that one time, but if not, you should maybe think of crafting a procedure.