Workout builder

[yokuha]

@david

You can use the “Import” button at the bottom of the dialog. You can also bulk import zwo, mrc and erg files into a folder in your training library:

Hi @David , thanks for the feedback. I wasn’t clear:-o
There don’t generally seem to be zwo files available for download for the Zwift workouts… Thus, what I get is a copy-and-paste version of the web info on whatsonzwift:

10min from 40 to 70% FTP
1x 3min @ 95rpm, 80% FTP,
1min @ 65rpm, 105% FTP
1min @ 85rpm, 65% FTP
1x 3min @ 100rpm, 80% FTP,
1min @ 65rpm, 105% FTP
1min @ 85rpm, 65% FTP
1x 3min @ 70rpm, 80% FTP,
1min @ 100rpm, 105% FTP
1min @ 85rpm, 65% FTP
1x 3min @ 95rpm, 80% FTP,
1min @ 65rpm, 105% FTP
2min @ 85rpm, 65% FTP
4min @ 95rpm, 90% FTP
4min @ 85rpm, 80% FTP
4min @ 75rpm, 70% FTP
4min @ 75rpm, 90% FTP
4min @ 85rpm, 80% FTP
4min @ 95rpm, 70% FTP
12min from 90 to 60% FTP
2min @ 65rpm, 75% FTP
1min @ 65rpm, 75% FTP
2min @ 95rpm, 75% FTP
1min @ 65rpm, 75% FTP
2min @ 65rpm, 75% FTP
6x 1min @ 100rpm, 75% FTP,
1min @ 60rpm, 75% FTP
3min @ 50% FTP

I’ll post the (short) Python code that I use to convert that to an i.icu-format specification when I am home tonight, but the easiest was, obviously, if I could paste this into an i.icu dialog and get the appropriate workout specs in i.icu;-)
The “trickiest” thing is the ‘,’ at end of line as a continuation symbol;-)

[david]

Where are you cut and pasting from on whatsonzwift? I just had a look there and couldn’t find any text representation of the workout.

[yokuha]

It’s this left “sidebar” – from this whatsonzwift page:

Btw., if you/i.icu could actually directly parse these website that would be cool;-))

[yokuha]

And this code will parse it into i.icu-compatible text

#!/usr/bin/env python
# -*- coding: utf-8; fill-column: 120; truncate-lines: t -*-

import re
import sys


def convert(zwift):
    icu = ""
    multiplier = False

    # expand multiplied lines
    for line in zwift:
        if re.match('^[1-9]+[xX] .*,\\s*\\Z', line):
            icu += '\n' + line[:3] + '\n'
            icu += '- ' + line[3:]
            multiplier = True
        elif re.match('^[1-9]+[xX]', line):
            icu += '\n' + line[:3] + '\n'
            icu += '- ' + line[3:]
            multiplier = False
        elif multiplier == True:
            icu += '- ' + line + '\n'
            multiplier = False
        else:
            icu += '- ' + line

    # replace words and fix grammar
    icu = icu.replace('min', 'm').replace('sec', 's') \
            .replace('from', 'ramp').replace(' to ', '-') \
            .replace('FTP', '').replace(',', '').replace('@', '').replace('\n+', '\n') + '\n'
    icu = re.sub('([0-9]+m) ([0-9]+s)', '\\1\\2', icu)

    return icu



if __name__ == '__main__':
    with open(sys.argv[1], 'rt') as zwift:
        print(convert(zwift))

[Max_Nelson]

http://cubingtest.byethost9.com/Other/CyclingCalculator.html

[Frederic_BATTIE]

Max_Nelson : thank you !!! this literaly mades my day

[david]

Thats very cool, tx for publishing it!

[yokuha]

Hey @Max_Nelson , nice for the community to put this up!

Then, it doesn’t really help me;-) Using Emacs & my Python code above gets me the i.icu text much faster than even opening yet another website – thus I would hope that @David will still eventually add an “WoZ importer/converter” to the workout builder of i.icu;-) That would be an real improvement (for me)

[Steve_Allinson]

It’s fabulous. Brilliant you. Thank you.

[app4g]

@david playing around w/ the builder. Noticed something strange.

for the last interval. I wanted to change it from 4m47s to 4m but noticed something interesting, when putting 4m00s instead of 4m, looks like it interpreted it differently, it considered it as not part of the 3x interval.

Nothing serious, just thot I’ll share

[Ben]

cheers for sharing, almost forgot I made this aha

[Max_Nelson]

Thanks for putting that together @Ben. I noticed you had some Rubik’s cube stuff too. I can’t solve them, but my 13 yo kid is crazy into them. Has every cube, pyramid, dodecahedron, etc imaginable and can solve up to a 10 by 10.

[Ben]

Good stuff, you should have a play with the 3x3. Using a beginner technique you’d be surprised how quickly you’ll be able to solve your first one.

David, I should have added, if you want to steal any code to integrate it into here just take it. It’s the only time I’ve ever used js so it’s probably junk could be a start

[david]

Tx I fixed that weirdness and will deploy Monday AM GMT+2.

[Brian_Chapman]

Have you thought about allowing workouts by HR zone? Example:

Warm Up

• 40m 1-2 HRZ

Hill Sprints: 10x

• 10s 7 HRZ
• 30s 1-1.5 HRZ, rest interval

Cool Down

• 8m 1-1.8 HRZ

In this example the HRZ would indicate to use your current heart rate zones. You could enter a fraction and it would calculate the equivalent HR for you.

[david]

Yes I do have that (and power zones) on the todo list but I didn’t think of doing it that way. Tx.

[app4g]

how would those workouts look like? As in how would it be able to be downloaded to be executed?
currently I’m pulling workouts as MRC files and parsing that.

[david]

Power based workouts using zones will still be downloadable as MRC, ERG or ZWO. The zones will be resolved at download time.

HR and pace based workouts can be uploaded to Garmin Connect for sync to Garmin devices. I will do a .fit file download for them at some point.

[app4g]

thanks. I was thinking along the lines of users who which to instead train using HR on the bike actually. sorry for not being more specific.

[yokuha]

I’ve updated my Python script to create i.icu-style workouts directly from the workout URL – run as zwift-to-i.icu https://whatsonzwift.com/workouts/build-me-up/week-7-lox to extract the workout from the whatsonzwift workout page and write out in i.icu format.

Now also retrieves the workout description from the website

#!/usr/bin/env python
# -*- coding: utf-8; fill-column: 120; truncate-lines: t -*-

import re
import sys

import requests
from bs4 import BeautifulSoup


def metadata(url):
    name = url.rpartition('/')[2]
    page = requests.get(url)
    soup = BeautifulSoup(page.content, "html.parser")
    results = soup.find(class_="two-thirds column workoutdescription").find_all('p')
    return results[-2].text


def workout(url):
    name = url.rpartition('/')[2]
    page = requests.get(url)
    soup = BeautifulSoup(page.content, "html.parser")
    results = soup.find(id=name).find_all(class_="textbar")
    zwift = ''
    for element in results:
        zwift += element.text.strip()
        zwift += '\n'
    return zwift[0:-1]


def convert(zwift):
    icu = ""
    multiplier = False

    # expand multipliers into old text-format multiple lines :-o)
    zwift = re.sub('FTP,', 'FTP,\n', zwift)
    # expand multiplied lines
    for line in zwift.split('\n'):
        if re.match('^[1-9]+[xX] .*,\\s*\\Z', line):
            icu += '\n' + line[:3] + '\n'
            icu += '- ' + line[3:] + '\n'
            multiplier = True
        elif re.match('^[1-9]+[xX]', line):
            icu += '\n' + line[:3] + '\n'
            icu += '- ' + line[3:] + '\n'
            multiplier = False
        elif multiplier == True:
            icu += '- ' + line + '\n\n'
            multiplier = False
        else:
            icu += '- ' + line + '\n'

    # replace words and fix grammar
    icu = icu.replace('min', 'm').replace('sec', 's') \
            .replace('from', 'ramp').replace(' to ', '-') \
            .replace('FTP', '').replace(',', '').replace('@', '').replace('  ', ' ') \
            .replace('\n+', '\n') + '\n'
    icu = re.sub('([0-9]+m) ([0-9]+s)', '\\1\\2', icu)

    return icu



if __name__ == '__main__':
    print(metadata(sys.argv[1]), '\n')
    print(convert(workout(sys.argv[1])))