Successive calls to Sentence() always return the (almost) same sentence
#9
Comments
|
For more details about my use case, this would allow me to do this: diff --git a/main.go b/main.go
index 31b7188..76a7048 100644
--- a/main.go
+++ b/main.go
@@ -116,13 +116,12 @@ func generateMessages(seed int64, senders int, limit int) []*Message {
source := rand.NewSource(seed)
random := rand.New(source)
lorem := loremipsum.NewWithSource(source)
- bodies := lorem.SentenceList(limit)
for i := 0; i < limit; i++ {
delay += random.Int63n(maxdelay)
msg := &Message{
Delay: time.Millisecond * time.Duration(delay),
From: random.Intn(senders),
- Body: bodies[i],
+ Body: lorem.Sentence(),
}
messages = append(messages, msg)
}And this would result in always the same sequence of random numbers and messages, even when changing the |
n-peugnet
changed the title
Sentence() always return the same sentenceSentence() always return the (almost) same sentence
n-peugnet
added a commit
to n-peugnet/loremipsum
that referenced
this issue
Jan 13, 2025
By calling li.shuffle() at the end of Sentence() instead of in SentenceList(), successive calls to Sentence() actually return very different sentences instead of small variations of the same sentence. Closes: go-loremipsum#9
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I expected successive calls
Sentence()to return (very) different sentences (with always the same sequence for a given seed), but it currently always returns (almost) the same one.I am forced to generate an array of sentences beforehand using
SentenceList(), but this is suboptimal, for multiple reasons:Sentence()in my loop, along the rest of my random number generation, and get a new sentence each time, I would not have this issue.I know that changing this might be a breaking change for some people, so I am not sure what would be the best way to implement it. I tested this patch:
And it behaves as I would like it to. But it would probably be better to add a new function, for example
NewSentence(), to keep the behaviour of the previous one intact, and maybe deprecating it.Also I didn't test the other methods for words and paragraphs, but they probably have the same issue.I tested
Word()andParagraph()and they didn't display the same issue, so maybe applying the patch and releasing a minor version would be enough?What do you think about this?
EDIT: Sorry, I tested this a bit more, and it turns out the sentences are slightly different, but not in the sense I was expecting. They are all a variation of the same sentence. I expected them to be very different.
For example, the following code...
...prints this:
Results for
Word()instead ofSentence()Results for
Paragraph()instead ofSentence()Whereas with my proposed patch, I get the following, which is what I expected:
(Results for
Word()andParagraph()are identical.)The text was updated successfully, but these errors were encountered: