plz help me in this code it goes in infinite loop.

p = int(input("Enter how many numbers you want to enter:"))
def bubbleSort(l):
for passnum in range(len(l)-1,0,-1):
for i in range(passnum):
if l[i]>l[i+1]:
temp = l[i]
l[i] = l[i+1]
l[i+1] = temp

l = [int(input("Enter the numbers :")) for _ in range(p)]
bubbleSort(l)
l.reverse()
max = l[0]
while(1):
for h in range(len(l)):
if(max%l[h] == 0):
print "LCM is:",max
break
max = max + 1
print "Thank you"

This will meet two different use cases that are not exactly a wiki - but add wiki like functionality to LMS systems:

• A large single document editable by the whole class where change tracking is important
• A collaborative editing document for small groups within a course where tracking is important

This is related to #82 in that i faced and figured this issue together, but i forgot to mention this in issue and/or the fix i pushed.

I will push a PR with any placeholder in description.

It would be really nice to see a TSUGI implementation (as a utility) added to composer (https://getcomposer.org/) for easy implementation into projects.

Right after cloning the repository and correctly setting up config.php, PHP gives an error:

unexpected end of file in /tsugi/lib/lms_lib.php
PHP Stack trace:, referer: http://localhost/tsugi/dev.php
PHP 1. {main}() /tsugi/dev.php:0, referer: http://localhost/tsugi/dev.php
PHP 2. require_once() /tsugi/dev.php:4, referer: http://localhost/tsugi/dev.php
PHP 3. require_once() /tsugi/config.php:99, referer: http://localhost/tsugi/dev.php

After removing lines 1025-1031 of lms_lib.php, the error is resolved.

@csev

I just upgraded to the latest version and now many of the links in the UI no longer work. This seems to be due to the fact that ".php" is missing from the URL's. For example, the upgrade link on the main page is "/admin/upgrade", and when I click it, gives a "not found" error.
Was there a change that required me to reconfigure something so that ".php" was inferred?

Something from ELMS:LN / Grav land. Grav abstracts all changes made after deploy into the user directory. This makes it super easy to know what to upgrade, what to version control, and where all credential data is vs what's safe to give out. ELMS:LN also does something similar using symlinks where all of our deploy specific changes actually live in a config directory. In practice what happens:
https://github.com/elmsln/elmsln-config-example is git cloned and replaces config folder in elmsln high level structure
https://github.com/elmsln/elmsln/blob/master/.gitignore#L5 ensures it's ignored from VC

Deploy UX that will never cause a data loss as a result:

cd /var/www/elmsln
git fetch --all
git reset --hard origin/master

Or when new version of our codebase release
cd /var/www/elmsln && git pull origin 0.10.x will safely work on any deploy that's ever made which doesn't modify this core convention. This allows us to start automating deployment and maintenance scripts.

This will never blow away anything made after the initial stamp down of the data. It's not to say my structure is right, just that I know it helped our developer UX immensely / simplified process after the fact. Grav's user folder offers similar workflow advantages when it comes to upgrading "core" at a later date in the same regard.

I am learning files in python. Actually in this I am reading a file in which extracting the lines which starts with From and then splitting the line into individual words. Then I am storing the line_split[1] into a list and then printing the list.
In output while printing the list I am getting each output twice. why?

name = input("Enter file:")
if len(name) < 1 : name = "mbox-short.txt"
handle = open(name)
count=0
d=list()
for line in handle:
if not line.startswith('From'):
continue
else:
line_split=line.split()
d.append(line_split[1])
#count+=1
print(d)
dct=dict()
for value in d:
dct[value]=dct.get(value,0)+1
maximum=max(dct.values())
final=[key for key,value in dct.items() if value==maximum]
print(key,maximum)
#print(count)'''

NOTE: This issue is more like an issue with canvas's implementation of LTI, but canvas being such a big player in the space, we need to work out to see what we can do to support their implementation in this regard.

Problem in canvas (iss, issuer):

Issue is, as described in the docs:

This requests contains an issuer identifier (iss) to recognize Canvas is launching the tool. As the issuer, Instructure-hosted Canvas intances all use the following, regardless of the specific account domain(s) that the tool was launched from:

https://canvas.instructure.com (Production environment launches)
https://canvas.beta.instructure.com (Beta environment launches)
https://canvas.test.instructure.com (Test environment launches)

And as a matter of fact, i recently did a canvas deployment, and on our xyz.com domain where canvas in hosted, we are getting https://canvas.instructure.com as issuer instead of xyz.com.
I know we can change the issuer to anything doing some config, but thats only us, we can't force everyone to set proper and unique issuer ids in their canvas systems. And also, instructor's hosted canvas instances would still have this same issuer despite whatever the domain/subdomain is.

Related links:
1- https://community.canvaslms.com/thread/36682-lti13-how-to-identify-clientid-and-deploymentid-on-launch#comment-153280
2- https://community.canvaslms.com/message/202373-issuer-identifieriss-issues-finding-iss-and-uniqueness-of-iss

Problem in TSUGI because of above:

Because of the uniqueness validation on issuer Id in TSUGI(first field at /admin/key/issuer-add form), we can really just add 1 canvas site as issuer, we can't really support multiple sites because no matter what the domain of the site is, with default configs, it will give iss(issuer id) as https://canvas.instructure.com.

I hope i have explained the problem correctly, if not let me know and i can re-phrase.

We want to work this out in tsugi codebase and share the solution with community. We are just looking for some hints as to what can be a good approach to fix this issue(in timely manner).

In peer-grader's code viewer, the real Line 1 is displayed as "Line 2". Fixing the location below may work?? Maybe we can move lines 150-154 all on the same line, in order to avoid any unwaned line-breaks inside the pre tag.
https://github.com/csev/tsugi/blob/master/mod/peer-grade/peer_util.php#L152

I was working on one assignment of PR4E. I had a dictionary which contains a string as its key and the number of times it occurs in a document is its value.

For example, myDict = {'My String':1, 'Another String': 2, 'Third String':7}.

I wanted to sort the dictionary on its values so I can get the key with max value. I used sorted function to do the job.

sorted(myDict.items(), key = lambda myDict: myDict[1], reverse=True)

When I run the script on my laptop it runs perfectly giving me the sorted (descending order) dictionary as list of tuples. But, on autograder, it somehow fails to sort the dictionary on values.

Expected Output after sorted(): [('Third String', 7), ('Another String', 2), ('My String', 1)]
Autograder Output after sorted(): [('Another String', 2), ('My String', 1), ('Third String', 7)]

What can be wrong here?

If you have a folder with spaces anywhere in the path - this type of error will happen.

Currently the minimum version for Tsugi is 5.5 - mostly because the first few versions of 7 were wonky. Because of a bug that was not fixed until 7.1.9, Tsugi has some wonky JavaScript in Output.php to compensate for the bugs in cookiless session support in PHP 7.

https://www.php.net/ChangeLog-7.php#7.1.9
https://bugs.php.net/bug.php?id=74892

While Tsugi does not currently have lots of 7-special syntax in the code base (perhaps none), it would be nice to remove this wonky workaround 7.0 / 7.1 code from three years ago. But we need folks to get to 7.2 or later before we remove the code.

I need help in this program.

After signing up for py4e I was pleasantly surprised to have been awarded an Open Badge. However, my Badgr Backpack doesn't accept it and unfortunately, the only message given is Badge import failed. The given badge failed validation: [object Object], [object Object].

My educated guess at this validation failure is that, after extracting the assertion data from the PNG using the hosted obi-sample and inspecting it, one issue may be that the URI for the assertion is not for the assertion, but for the unbaked image. The line that adds that URI is

I understand that awarding badges may not be core to TSUGI (looking at earlier issues csev/obi-sample#2 and csev/obi-sample#3 as well), but I couldn't find an issue in this repository and did want to mention the incompatibility. If awarding Open Badges is indeed not part of TSUGI anymore, maybe documentation could point at LMS-asserted badges, awarded by the LMS itself (like Brightspace appears to do) or through external issuers like Credly or Badgr (like Canvas and Moodle do AFAIK).

Is this a bug? Should it be path(2048)?

My admin/upgrade output below

Checking plugins table...
Checking for any needed upgrades...
Checking admin/lti/database.php ...
-- Checking table lti_key
-- Checking table lti_user
-- Checking table lti_context
-- Checking table lti_link
-- Checking table lti_link_activity
-- Checking table lti_membership
-- Checking table lti_event
-- Checking table lti_link_user_activity
-- Checking table lti_service
-- Checking table lti_result
-- Checking table lti_nonce
-- Checking table lti_domain
-- Checking table tsugi_string
-- Checking table profile
-- Checking table cal_event
-- Checking table cal_key
-- Checking table cal_context
-- Checking upgrade admin/lti/database.php 
-- Current data model version 201801271430 
Checking admin/key/database.php ...
-- Checking table key_request
-- Checking upgrade admin/key/database.php 
-- Current data model version 2014042200 
Checking admin/blob/database.php ...
-- Checking table blob_file
-- Creating table blob_file
Unable to create create table blob_file ( file_id INTEGER NOT NULL KEY AUTO_INCREMENT, file_sha256 CHAR(64) NOT NULL, context_id INTEGER NULL, link_id INTEGER NULL, file_name VARCHAR(2048), deleted TINYINT(1), contenttype VARCHAR(256) NULL, path VARCHAR(2048) NULL, content LONGBLOB NULL, blob_id INTEGER, json TEXT NULL, created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP, accessed_at TIMESTAMP NOT NULL DEFAULT '1970-01-02 00:00:00', INDEX `blob_indx_1` USING HASH (`file_sha256`), INDEX `blob_indx_2` (`path(256)`), CONSTRAINT `blob_ibfk_1` FOREIGN KEY (`context_id`) REFERENCES `lti_context` (`context_id`) ON DELETE SET NULL ON UPDATE CASCADE ) ENGINE = InnoDB DEFAULT CHARSET=utf8 42000:42000:SQLSTATE[42000]: Syntax error or access violation: 1072 Key column 'path(256)' doesn't exist in table
create table blob_file ( file_id INTEGER NOT NULL KEY AUTO_INCREMENT, file_sha256 CHAR(64) NOT NULL, context_id INTEGER NULL, link_id INTEGER NULL, file_name VARCHAR(2048), deleted TINYINT(1), contenttype VARCHAR(256) NULL, path VARCHAR(2048) NULL, content LONGBLOB NULL, blob_id INTEGER, json TEXT NULL, created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP, accessed_at TIMESTAMP NOT NULL DEFAULT '1970-01-02 00:00:00', INDEX `blob_indx_1` USING HASH (`file_sha256`), INDEX `blob_indx_2` (`path(256)`), CONSTRAINT `blob_ibfk_1` FOREIGN KEY (`context_id`) REFERENCES `lti_context` (`context_id`) ON DELETE SET NULL ON UPDATE CASCADE ) ENGINE = InnoDB DEFAULT CHARSET=utf8

"You need to set $CFG->google_client_id in order to use Google's Login" -- Beyond the "we don't feel like supporting login systems" message it gives, is there a real reason? Otherwise this would be very difficult to get into certain environments unless this is only the requirement for the admin user to get logged in (tho also odd since I had to put in a password into a file manually in order to use this)

In order to better service multi-tenant SASS installations and to better handle LTI 1.1 -> LTI Advantage transition, the lti13_ fields need to move from the lti_key table and into a new lti_issuer table with a many-to-one relationship between keys and issuers.

Hi, I'm a new user of Tsugi and I noticed that this project has the vendor directory included in the repository. I recommend it should be removed and added to the .gitignore file, per the best practices described in "Should I commit the dependencies in my vendor directory?" I can make a new PR with the changes and update the docs if maintainers want this.

Is there some reasoning I'm missing for including it in the repository?

largest = 0
smallest = 0

while True:
num = input("Enter a number: ")
if num == "done" :
break
try:
num == float (num)
except:
print ("Invalid input")

#############

if num > largest:
num is largest
print("Maximum is num")

if num < smallest:
num is smallest
print("Minimum is smallest")

*Im not sure but I have been working on this for awhile. I need the result to read:
Invalid input
Maximum is 10
Minimum is 2

ame = raw_input("Enter file name: ")

fh = open(fname)
for i in fh:
fh = i.upper()
print fh.rstrip()

def computepay(hours,rate):
if h > 40:
reg = hours * rate
otp = (hours - 40) * (rate * 0.5)
pay = reg + otp
else:
pay = hours * rate
return pay

hrs = input("Enter Hours:")
rte = input("Enter Rate:")
h = float(hrs)
r = float(rte)
xp = computepay(h, r)
print("Pay",xp)
#what is the wrong with may program. everything is correct put the autograder does not give approve it!!!!?

This will be wrapping the Java Tsugi in SCALA APIs. It should not be too difficult.

The format number (e.g. 42,27) makes the default $lat and $lng useless when hosting outside US.
Could be a workaround adding on
https://github.com/csev/tsugi/blob/master/mod/map/index.php#L75

It seems like configuration settings are more or less thrown in as people think of them into config-dist.php. Would it possibly be easier to read as well as modify and put into version control if this was abstracted to a yaml or json file that config then loads in? This would also make it easier to write a CLI / YO that could write the config options automatically for people increasing devUX down stream

In Context::getGradeToken - we sent the subject as key_key on token requests - key_key was a Tsugi-made up string that was never part of the security contract prior to the issuer refactor. This passed the cert suite and interacted with all the LMSs successfully this way. Post the issuer refactor - this looks like it really means to be the client_id. Make sure after this change is tested to go into Context.php and cleanup all the TODOs.

Exploring my canvas installation logs, it was complaining about missing keys in public_jwk, missing keys were alg and use to be specific.
What works though is if i add these keys as per given from https://tsugi.xyz.org/lti/keyset?issuer=https%3A%2F%2Fcanvas.instructure.com end point.

So i think these missing keys should be added in the "Canvas Configuration URL (json)" generated at this endpoint https://tsugi.xyz.org/lti/store/canvas-config.json?issuer=https%3A%2F%2Fcanvas.instructure.com

Prior to January 2017, Tsugi's login.php made an incorrect assumption about the "GUID" returned for a user from Google upon successful login. The assumption was that this was a persistent identifier - which turns out not to be the case. So every month or so, a user might get an additional profile but not an extra user record. In January 2017, code was added to use the email as the logical key in the profile instead of the user "ID" from Google and to use only the most recently created profile and ignore the earlier profiles.

Enough time has passed, that there is a need to teach "Upgrade Database" to clear out the old entries.

Since LMS vendors are walking away from LTI 2.0 in the next 12 months, I will get a head start and remove LTI 2.0 support from Tsugi. If this is a problem for anyone - please let me know. I am assuming that no one every used Tsugi LTI 2.0 except to test Sakai, Canvas and Blackboard.

I will snap a release and branch before I drop all that code - but it won't be long unless I hear from folks really soon.

f string literals won't work in https://www.py4e.com/tools

https://www.python.org/dev/peps/pep-0498/

I'd like to upgrade Tsugi to use Bootstrap version 4.5.0. This version is a major rewrite from v3, so many aspects of Tsugi will need to be updated. https://getbootstrap.com/docs/4.5/migration/

I'm not sure the best way to avoid breaking Tsugi tools as they may rely on or assume that Bootstrap 3 is included and piggy-back off of the styling. Any suggestions are welcome.

Right now, database.php is only called from admin/upgrade.php - but for simpler development - we should make it so that the database.php can be run directly when in developer mode.

Line 198 of mod/php-intro/a05.php (https://github.com/csev/tsugi/blob/master/mod/php-intro/a05.php#L198) should have been surrounded by a try-catch block as well.

A student submitted a page with a hyper link with text "Log out" instead of "Logout". $crawler->selectLink('Logout') returns an empty list that causes an uncatched exception. This resulted in an uninterpretable error message returned to the auto-grader interface.

Lots of servers are configured with some kind of a proxy layer that does not pass the "Authorization" header. This might be automatic checking of basicauth or perhaps just overly conservative default configurations. The (quite non-standard) approach I have taken is to replicate the "Authorization:" header in the "X-Oauth1-Authorization:" header. Nothing different - just send both headers with identical information. Incoming OAuth 1.0a protected web services can look for "X-Oauth1-Authorization" when "Authorization" is not present.

Some may criticise this but I see no alternative. Hosting providers like 1and1 simply will never listen to someone who begs to configure "Authorization" to be passed into Apache. Ah well.

This has no effect on POST data - just header data in web service requests.

Comments welcome.

If $CFG-dbprefix is set and columns need to be added, the table name is wrong in th eupgrade SQL, because the $CFG->dbprefix is prepended twice (at least in two locations)

admin/lti/database.php cf lines 690-704: $CFG->dbprefix is added on line 696 and on 700. As a test removed it on line 700.

Similar issue in admin/blob/database.php lines 95 - 109

https://www.py4e.com/
Autograder Ex 3.1

In this sentence:"Pay the hourly rate for the hours up to 40 and 1.5 times the hourly rate for all hours worked above 40 hours."
(1.5 is wrong) to get the 498.75 as requested you must multiply by 0.5 times and not 1.5.

Dear dev,
If you are echoing into JS inside a HTML document (as you are in your example), you risk opening up a huge XSS vulnerability.

echo('            context_title: '.json_encode($CONTEXT->title).",\n");
...

you must use "JSON_HEX_TAG" in the json_encode function.
instead of above code you must use below:

echo('            context_title: '.json_encode($CONTEXT->title).",\n",JSON_HEX_TAG);
...

https://security.stackexchange.com/questions/175630/passing-php-code-directly-into-javascript-in-html5
https://stackoverflow.com/questions/20942452/why-does-script-cause-a-dom-tree-break-on-the-browser
https://security.stackexchange.com/questions/110101/proper-way-to-protect-against-xss-when-output-is-directly-into-js-not-html

In the past few months, the LTI Working Group has added a lot of detail to the LTI 1.1 to LTI Advantage migration strategy. The approach used in Sakai-19 reflected the strategy of about 3 months ago which has been changed and improved through the WG.

https://jira.sakaiproject.org/browse/SAK-41781

This JIRA will be the modifications to align with the transition guide:

https://www.imsglobal.org/spec/lti/v1p3/migr

I am trying to use Tsugi console but if I put in the custom key and secret that my lti provider gave me, every time it goes back to '12345' in the key field.

I tried on local instance and https://lti-tools.dr-chuck.com/tsugi/dev.php and got the same error.

When logged in, "Badge" link on main page goes to https://www.py4e.com/tsugi/badges.php which doesn’t seem to exist. [Not sure if this is a tsugi issue.]

I believe some work was done here to support dark mode. The default should be the mode specified by the browser.

Most browsers supply this media query https://caniuse.com/#search=prefers-color-scheme
This can also be read in javascript. This article has the best practice ways to support this: https://web.dev/prefers-color-scheme/#finding-out-if-dark-mode-is-supported-by-the-browser

Hello everyone
am new to python and coding and programming in general but want to learn more but i dont reall know the rith part to follow.
i need some guidance

Issue

On/admin/key/keys there are some keys that come installed by default. I wanted to remove them so I did.

However, after removing all the keys, it blocks the creation of a new key

?php if ( count($newrows) < 1 ) { ?>
<p>
You have no Tenant keys for this system.
</p>
<?php } else {
    $extra_buttons = array(
        "Insert New Key" => "key-add"
    );
    Table::pagedTable($newrows, $searchfields, false, "key-detail", false, $extra_buttons);
}
if ( isAdmin() ) { ?>
<?php } ?>

excepted result

You should always be able to add keys even if you have no keys in your database.

This is something that is waiting for someone with interest and willingness to put the resulting code in production.