{Ignore/Remove Redundancy} + {Condition for Directory}

[alfa-147]

Since LongPathsEnabled only allows the total path length to exceed 260 characters, not each individual segment (in my case the "directories"), I would like to ask if these two workarounds to the problem are feasible and, if so, how.
But first, a quick reminder to my settings (they're essentially the same as my previous posts with a few changes):
Directory command:
"directory": ["\fM C:/GDL/_py/PROCESS.py:main", "{tags_copyright[0]:?//}", "{tags_copyright:?//Sa/}", "{tags_artist[0]:?//}"],

Directories structure:

tags_copyright[0]
   └── tags_copyright
              └── tags_artist
                       └── FILE

Python script part for replace function: (I reduced the mention of the replacement list to a single point for a faster explanation)

REPLACEMENTS = {
    "mori_calliope_ch."  : "HOLOLIVE"
     }
def transform(data, func, src, dst=None):
    if dst is None:
        dst = src

    value = data.get(src)
    if not value:
        data[dst] = ""
    elif isinstance(value, (list, tuple)):
        data[dst] = [func(ele) for ele in value]
    else:
        data[dst] = func(value)


def repl(v):
    return REPLACEMENTS.get(v, v)

def replace(metadata):
    transform(metadata, repl, "tags_copyright")
    return ""

Workaround n°1: Ignore "dst" value redundancy
Applying these settings may cause directories to have the same substitution value (dst) repeated multiple times, causing the directory name to approach its limits.
e.g.
tag_copyright post:

hololive
hololive_english
mori_calliope_ch.
nerissa_ravencroft_ch.
takanashi_kiara_ch.

The directory will be named: "HOLOLIVE\ HOLOLIVE - HOLOLIVE - HOLOLIVE - HOLOLIVE - HOLOLIVE"
Question: Is it possible to have the script ignore/remove any redundancy of already used dst values?
*I also consider it a success to simply replace the redundant values ​​with a single character.

Workaround n°2: using directory option with conditionals
Question: Is it possible to predetermine the length of each individual path segment with an f-string or something, so that another, more simplified path (limited only to the tag_copyright[0]) is used if its length exceeds the allowed length?
*I have found this post on stackoverflow which mention this possibility with [str.split()] and [len(directory)] but I don't know, maybe it can't be the solution I'm looking for.
https://stackoverflow.com/questions/32172659/determine-if-a-directory-name-has-more-than-260-characters

In any case, Thanks in advance for any advice.

[alfa-147]

I apologize for the annoying situation but in the meantime I managed to solve only the second part of the post by myself. Hoping to not disturb you too much, I really need your help to solve this tedious situation. Thanks.

[mikf]

I would suggest building all of a directory name in a specialized Python function, or at least assigning the entire directory name to a single metadata value, so that you can control its length. Either in Python or in a format string with {FIELD_NAME[b:250]}

To get rid of duplicate entries, convert the tag list to a dict after replacing its entries and before combining/using them. Maybe convert it back to a list to be able to access individual elements by index:

tags = [
    "hololive",
    "hololive",
    "hololive_english",
    "mori_calliope_ch.",
    "hololive",
    "takanashi_kiara_ch.",
    "hololive",
]

tags_dict = dict.fromkeys(tags)
tags_uniq = list(tags_dict)

for t in tags_uniq:
    print(t)

[alfa-147]

First of all I would like to thank you for the help you have given me, however I am having some difficulty in applying it.
I created the dict tag list and placed it after the REPLACEMENTS part in the script just like you said, but I have no idea where to place the rest. I tried placing them under def transform, def repl(v) and def replace(metadata) or even just placing them after the dict, but each time the situation remains the same producing duplicate entries.
I'm sorry, could you give me some further enlightenment on how to proceed?

[alfa-147]

I found out the reason why it doesn't seem to work.
Currently the dictionary prevents duplicate entries (keys) and not its outputs (values).
When downloading, after replacing keys within the metadata with my own values, how can I ensure to prevent the duplicate output values to be considered, since each post already has unique keys?
Explanation:

{keys}                                   {value}
    "hololive"                  :            "HOLOLIVE",
    "hololive_english"          :            "HOLOLIVE",
    "mori_calliope_ch."         :            "HOLOLIVE",
    "takanashi_kiara_ch."       :            "HOLOLIVE",

If a post has all these keys in metadata, the directory output is: "HOLOLIVE HOLOLIVE HOLOLIVE HOLOLIVE".
Avoided the duplicacy: "HOLOLIVE".

[alfa-147]

Current Status Update
I've made some improvements in terms of efficiency: now instead of repeating the value for each key inside the dictionary, all keys point to a unique value.

# Before:
REPLACEMENTS = {
    "hololive"                  : "𝐇𝐎𝐋𝐎𝐋𝐈𝐕𝐄",
    "hololive_english"          : "𝐇𝐎𝐋𝐎𝐋𝐈𝐕𝐄",
    "mori_calliope_ch."         : "𝐇𝐎𝐋𝐎𝐋𝐈𝐕𝐄",
    "takanashi_kiara_ch."       : "𝐇𝐎𝐋𝐎𝐋𝐈𝐕𝐄",
}

# Now:
REPLACEMENTS = dict.fromkeys([
    "hololive",
    "hololive_english",
    "mori_calliope_ch.",
    "takanashi_kiara_ch.",
	    ], "𝐇𝐎𝐋𝐎𝐋𝐈𝐕𝐄"
    )

No progress for duplicate values, however I believe that the "fault" is due to:

def repl(v):
    return REPLACEMENTS.get(v, v)

because the function returns the value "𝐇𝐎𝐋𝐎𝐋𝐈𝐕𝐄" for every matching unique key in the dictionary, and since they all point to the same value, duplication occurs.
Knowing this, I tried to replace it using set() and list() methods, but always ended in errors like "TypeError: unhashable type: 'dict'", "TypeError: unhashable type: 'list'" or argument not defined.
I've been looking at a lot of guides and tutorials, but I'm missing something.
Sorry to have to rely on you again, but these are my limits in understanding Python, do you know where I'm going wrong?

[mikf]

You need to apply repl/REPLACEMENTS to all relevant tags first and then do the dict.fromkeys step.

I've slightly modified the transform function from #7183 (comment) here to filter out duplicates:

REPLACEMENTS = {
    "mori_calliope_ch."  : "HOLOLIVE",
}

def transform(data, func, src, dst=None):
    if dst is None:
        dst = src

    value = data.get(src)
    if not value:
        data[dst] = ""
    elif isinstance(value, (list, tuple)):
        uniq = dict.fromkeys(func(e) for e in value)
        data[dst] = list(uniq)
    else:
        data[dst] = func(value)


def repl(v):
    return REPLACEMENTS.get(v, v)

def replace(metadata):
    transform(metadata, repl, "tags_copyright")
    return ""

[alfa-147]

👍 Thank you so much! It works perfectly.