Possibly unintended pause

[basvdl97]

Here is an overview of two function in selenium_actions.py. I believe there is an unintended pause in the write_sentence function, claiming to pause between each character, where in fact it pauses after a word.

However ther is already a pause after a word, namely, at the end of the write_word function.

    def write_sentence(self, sentence):
        self.actions.pause(std_positive(1.3, 1, 0.2)) # Opening a sentence
        words = sentence.split(" ")
        if len(words) > 0:
            for i in range(len(words)-1):
                self.write_word(words[i])
                self.actions.pause(std_positive(0.21, 0.03, 0.011)) # Pauze between characters (within a word)
                self.write_character(" ")
            self.write_word(words[-1])

    def write_word(self, word):
        self.actions.pause(std_positive(0.47, 0.21, 0.05)) # Opening a word
        characters = list(word)
        if len(characters) > 0:
            for i in range(len(characters)-1):
                self.write_character(characters[i])
                self.actions.pause(std_positive(0.21, 0.03, 0.011)) # Pauze between characters (within a word)
            self.write_character(characters[-1])
        self.actions.pause(std_positive(0.2, 0.08, 0.01)) # Closing a word

@droefs Am I correct to assume that the pause in write_sentence preceding the comment # Pauze between characters (within a word) can be removed?

[basvdl97]

I found a possible second case where this occurs:

def key_down(self, value, element=None, addDelayAfter=True):
        if element is not None:
            self.click(element)
        self.actions.key_down(value)
        if addDelayAfter:
            self.actions.pause(std_positive(0.06, 0.035, 0.013)) # Specific version of addDelayAfterAction()
        return self

def write_character(self, character):
        special_characters = "!@#$%^&*()_+{}|:<>?"
        if character.isupper() or character in special_characters:
            self.actions.key_down("\ue008")
            self.actions.pause(std_positive(0.06, 0.035, 0.005)) # Time after shift press
        self.actions.key_down(character)
        self.actions.pause(std_positive(0.06, 0.035, 0.013)) # Dwell time
        self.actions.key_up(character)
        if character.isupper() or character in special_characters:
            self.actions.pause(std_positive(0.03, 0.015, 0.003)) # Time before shift release
            self.actions.key_up("\ue008")

In this function write_character there is a pause commented as # Dwell delay. I understand the need for the dwell however the timings seem to be exactly the same as the pause after key_down when addDelayAfter=True. Since addDelayAfter=True by default I believe there might be on average twice the delay that is intended.

[droefs]

Great catch! You are correct in both cases.

@droefs Am I correct to assume that the pause in write_sentence preceding the comment # Pauze between characters (within a word) can be removed?

Yes

Since addDelayAfter=True by default I believe there might be on average twice the delay that is intended.

Indeed. The argument addDelayAfter should be set to false when calling key_down(). This is probably a regression bug form when the optional addDelayAfter parameter was introduced. It also appears in the click() function and maybe also in others. I will add a addDelayAfter=False argument in all relevant functions.