ganeti

Ganeti.Utils

Description

Utility functions.

Synopsis

Debug functions

debug :: Show a => a -> aSource

To be used only for debugging, breaks referential integrity.

debugFn :: Show b => (a -> b) -> a -> aSource

Displays a modified form of the second parameter before returning it.

debugXy :: Show a => a -> b -> bSource

Show the first parameter before returning the second one.

Miscellaneous

applyIf :: Bool -> (a -> a) -> a -> aSource

Apply the function if condition holds, otherwise use default value.

commaJoin :: [String] -> StringSource

Comma-join a string list.

sepSplit :: Eq a => a -> [a] -> [[a]]Source

Split a list on a separator and return an array.

plural :: Int -> String -> String -> StringSource

Simple pluralize helper

ensureQuoted :: String -> StringSource

Ensure a value is quoted if needed.

Mathematical functions

stdDev :: [Double] -> DoubleSource

Standard deviation function.

data Statistics Source

Abstract type of statistical accumulations. They behave as if the given statistics were computed on the list of values, but they allow a potentially more efficient update of a given value.

Constructors

 SumStatistics Double StdDevStatistics Double Double Double

Instances

 Show Statistics

getSumStatistics :: [Double] -> StatisticsSource

Get a statistics that sums up the values.

getStdDevStatistics :: [Double] -> StatisticsSource

Get a statistics for the standard deviation.

getStatisticValue :: Statistics -> DoubleSource

Obtain the value of a statistics.

updateStatistics :: Statistics -> (Double, Double) -> StatisticsSource

In a given statistics replace on value by another. This will only give meaningful results, if the original value was actually part of the statistics.

Logical functions

Arguments

 :: Bool condition -> a "then" result -> a "else" result -> a "then" or else result depending on the condition

"if" as a function, rather than as syntactic sugar.

Parsing utility functions

parseChoices :: (Monad m, Read a) => String -> String -> [(a, String)] -> m aSource

Safe `read` function returning data encapsulated in a Result.

formatTable :: [[String]] -> [Bool] -> [[String]]Source

Format a table of strings to maintain consistent length.

printTable :: String -> [String] -> [[String]] -> [Bool] -> StringSource

Constructs a printable table from given header and rows

parseUnitValue :: Monad m => String -> m RationalSource

parseUnit :: (Monad m, Integral a, Read a) => String -> m aSource

Tries to extract number and scale from the given string.

Input must be in the format NUMBER+ SPACE* [UNIT]. If no unit is specified, it defaults to MiB. Return value is always an integral value in MiB.

exitIfBad :: String -> Result a -> IO aSource

Unwraps a `Result`, exiting the program if it is a `Bad` value, otherwise returning the actual contained value.

exitErr :: String -> IO aSource

Exits immediately with an error message.

exitWhen :: Bool -> String -> IO ()Source

Exits with an error message if the given boolean condition if true.

exitUnless :: Bool -> String -> IO ()Source

Exits with an error message unless the given boolean condition if true, the opposite of `exitWhen`.

logWarningIfBad :: String -> a -> Result a -> IO aSource

Unwraps a `Result`, logging a warning message and then returning a default value if it is a `Bad` value, otherwise returning the actual contained value.

warn :: String -> IO ()Source

Print a warning, but do not exit.

extractKey :: [Either Integer String] -> String -> ([Either Integer String], String)Source

niceSort :: [String] -> [String]Source

Sort a list of strings based on digit and non-digit groupings.

Given a list of names `[a1, a10, a11, a2]` this function will sort the list in the logical order `[a1, a2, a10, a11]`.

The sort algorithm breaks each name in groups of either only-digits or no-digits, and sorts based on each group.

Internally, this is not implemented via regexes (like the Python version), but via actual splitting of the string in sequences of either digits or everything else, and converting the digit sequences in Left Integer and the non-digit ones in Right String, at which point sorting becomes trivial due to the built-in `Either` ordering; we only need one extra step of dropping the key at the end.

niceSortKey :: (a -> String) -> [a] -> [a]Source

Key-version of `niceSort`. We use `sortBy` and `compare on fst` since we don't want to add an ordering constraint on the a type, hence the need to only compare the first element of the (key, a) tuple.

rStripSpace :: String -> StringSource

Strip space characthers (including newline). As this is expensive, should only be run on small strings.

newUUID :: IO StringSource

Returns a random UUID. This is a Linux-specific method as it uses the /proc filesystem.

getCurrentTime :: IO IntegerSource

Returns the current time as an `Integer` representing the number of seconds from the Unix epoch.

getCurrentTimeUSec :: IO IntegerSource

Returns the current time as an `Integer` representing the number of microseconds from the Unix epoch (hence the need for `Integer`).

clockTimeToString :: ClockTime -> StringSource

Convert a ClockTime into a (seconds-only) timestamp.

chompPrefix :: String -> String -> Maybe StringSource

Strip a prefix from a string, allowing the last character of the prefix (which is assumed to be a separator) to be absent from the string if the string terminates there.

>>> chompPrefix "foo:bar:" "a:b:c" Nothing

>>> chompPrefix "foo:bar:" "foo:bar:baz" Just "baz"

>>> chompPrefix "foo:bar:" "foo:bar:" Just ""

>>> chompPrefix "foo:bar:" "foo:bar" Just ""

>>> chompPrefix "foo:bar:" "foo:barbaz" Nothing

Arguments

 :: Int maxWidth -> String string that needs wrapping -> [String] string "broken" in lines

Breaks a string in lines with length <= maxWidth.

NOTE: The split is OK if:

• It doesn't break a word, i.e. the next line begins with space (`isSpace . head \$ rest`) or the current line ends with space (`null revExtra`);
• It breaks a very big word that doesn't fit anyway (`null revLine`).

trim :: String -> StringSource

Removes surrounding whitespace. Should only be used in small strings.

defaultHead :: a -> [a] -> aSource

A safer head version, with a default value.

exitIfEmpty :: String -> [a] -> IO aSource

A `head` version in the I/O monad, for validating parameters without which we cannot continue.

splitEithers :: [Either a b] -> ([a], [b], [Bool])Source

Split an `Either` list into two separate lists (containing the `Left` and `Right` elements, plus a "trail" list that allows recombination later.

This is splitter; for recombination, look at `recombineEithers`. The sum of "left" and "right" lists should be equal to the original list length, and the trail list should be the same length as well. The entries in the resulting lists are reversed in comparison with the original list.

recombineEithers :: (Show a, Show b) => [a] -> [b] -> [Bool] -> Result [Either a b]Source

Recombines two "left" and "right" lists using a "trail" list into a single `Either` list.

This is the counterpart to `splitEithers`. It does the opposite transformation, and the output list will be the reverse of the input lists. Since `splitEithers` also reverses the lists, calling these together will result in the original list.

Mismatches in the structure of the lists (e.g. inconsistent lengths) are represented via `Bad`; normally this function should not fail, if lists are passed as generated by `splitEithers`.