Description

When citing works by multiple authors the "and" separator in different languages leads the program to conclude those are separate citations. This could occur e.g. in the English abstract of a croatian article.

Expected behavior

"A and B, 2000" and "A i B, 2000" should be logged as one citation. Preferably "A i B, 2000" since Croatian matches mean the article is probably in Croatian.

Observed behavior

Both "A and B, 2000" and "A i B, 2000" get recorded as separate citations.

Description

Names with the "ç" don't get recorded as parts of citations.

Expected behavior:

The text includes the following sentence: "Mendonça i suradnici (2009) utvrdili su..."
The Excel file should contain "Mendonça i sur. 2009"

Actual behavior:

Excel contains "sur. 2009"

Description

If any .txt file is loaded, a warning pops up that ANSI encoding is preferred.

Suggestion

If the encoding of the file can be determined, show the warning only if it is not already ANSI.

Description

In the sentence Cohen's (1999) seminal paper..., an article by Cohen (1999) is cited and should be recorded as such. Currently, the citation appears as Cohen's 1999, instead of Cohen 1999.
This could also lead to duplicate citations being recorded if both the possessive and regular form occur.

The software does not pick up unpublished papers cited as Author(s) (in press) or Author(s) (u tisku - in Croatian).

Description

Certain phrases in the format of word, long number get recognised as citations, e.g. Washington, DC, 10694(000).

Expected behavior

The mentioned text doesn't get written in Excel.

Current behavior

The mentioned text returns a citation "DC 1069"

Suggested solutions

• Filter citations starting with "DC" as excluded
• Alter the part that catches years so that it must include 4 digits, no more, no less.

Description

Currently, no text in footnotes is analyzed. So, if it contains additional citations, they won't be recorded in the output file.

Description

The program closes after finishing gathering citations for a given file. Offer an option to analyze another file instead of closing.

Description

Some texts contain lots of single words followed by a string of numbers, which falsely get detected as authors.
Limiting first authors to only uppercase words would avoid this issue. This applies to solo authored works, pairs, and threes.

Expected behavior:

The text includes the following sentence: "Review of psychology (1234-5678-1234)..."
The Excel file should not contain anything from this sentence.

Actual behavior:

Excel contains "psychology 1234"

Potential issues

Misspeling author names will prevent the program from recognising them.

.Csv should consist of two columns, narrower and wider citations, with no empty columns provided.

Description

Croatian

Authors can be reffered to in different cases in Croatian (padeži). For example, Cohen, Cohenu, Cohenom. Possesive forms also alter the words - Cohenov. These all refer to the same source.

English

Cohen's (1999) paper is the same as the one referenced by Cohen (1999).

Suggested feature

Recognise different word endings in citations. If two citations differ only in the last letter, keep only one. Same for citations ending in 's.

Example

Superov (1999) članak potaknuo je daljnju diskusiju. (...) Na kraju, nitko nije imao bolju ideju od Supera (1999).
With recognising case, the citation in the Excel file would be a single Super 1999.

Potential issues

• Which ones to keep? Have a hierarchy (nominativ, genitiv...) and keep the highest one.
• Could this potentially confuse authors whose surnames actually differs in the last letter?
  • Make this an optional feature.

Grammatical cases implemented:

• recognise Croatian cases
• recognise Croatian possesives
• recognise English possesives
  • probably the easiest

Description

Right now the program detects multiple sources when they are listed directly after one another. However, they are recorded in one cell, and possibly duplicate other citations when they reference just one of the years listed.
A function that separates such citations into individual ones would help promote readability and reduce duplicated sources. The sources should be listed as such in the reference list.

Expected behavior:

The text includes the following sentence: "AuthorA (2000; 2002; 2003) thoroughly explored..."
The Excel file should list "AuthorA 2000", "AuthorA 2002" and "AuthorA 2003" as separate citations.

Actual behavior:

Excel contains a citation "AuthorA 2000 2002 2003"

This requires #40 to be implemented first, because clipboard text will have no default filepath to save to.

Current status

Individual characters that are "allowed" are listed in the program.
Omission of characters such as æ, ø, å...

Suggestion

Import a list of Unicode characters (that doesn't include numbers, special symbols...) and use that for maximum coverage.